We present MiHazeFree3D , a system that leverages millimeter-wave (mmWave) radar signals to predict 3D bounding boxes of vehicles and pedestrians in real-world traffic scenarios. While current 3D object detection methods rely primarily on cameras and LiDARs, their performance degrades significantly in rain, fog, or poor lighting conditions. Our system exploits mmWave radar’s ability to operate reliably in these challenging conditions, offering a complement to existing sensors without increasing computational costs. The key challenge in using mmWave for 3D detection lies in handling motion-induced errors and the specular reflection of mmWave signals. To address these issues, we developed a deep learning architecture with multiple feature fusion layers and trained it on diverse real-world scenarios. We evaluated MiHazeFree3D using data collected from mmWave radars mounted on the dashboard of an ego-vehicle driving through urban environments. Our results show that MiHazeFree3D detects and bounds both vehicles and pedestrians in tested conditions, including fog and low-light scenarios, highlighting the potential of mmWave radar for 3D object detection in autonomous driving systems.

Three-dimensional medical image segmentation plays a significant role in clinical diagnosis, treatment planning, and disease research, as it provides doctors with precise anatomical and lesion information and improves the accuracy and efficiency of medical decision-making. However, most existing 3D segmentation approaches rely heavily on densely volumetric data and often fail to perform segmentation properly for incomplete 3D volume acquisition, i.e., missing slices. In this work, we present InterFrameNet, a framework designed to predict intermediate lesion structures by modeling spatial relationships across frames, enabling robust segmentation performance under sparse acquisition conditions, without requiring full-volume information. Our method explicitly models cross-frame spatial continuity and leverages structural relationships between available frames to accurately infer missing lesion regions. This design significantly reduces the dependence on consecutive frames while fully exploiting contextual anatomical information. Extensive experiments on brain lesion datasets demonstrate that our approach achieves robust segmentation performance under sparse acquisition settings, offering a practical solution to maximize usability of incomplete clinical imaging data.

Intracerebral hemorrhage (ICH) is a life-threatening stroke that requires precise segmentation for effective treatment. To aid in diagnosing ICH, various deep learning-based methods have been proposed. However, these methods face challenges posed by irregular patterns and low-contrast boundaries in ICH images. In this paper, we introduce a novel conditional diffusion-based segmentation approach for ICH segmentation. Our framework leverages ResNet18 and a Transformer Block to enhance conditional feature extraction, as well as cross-attention mechanisms to align global and local features between the conditional and diffusion branches. Experiment results on the Instance2022 dataset show an improvement exceeding 10% in the Dice coefficient and IoU score, compared to the baseline diffusion-based model, MedSegDiffv2. More impressively, more than 20% improvement is achieved in terms of both the Dice coefficient and IoU score in cross-dataset evaluations on two benchmarks, i.e., the BHSD and PhysioNet, demonstrating excellent generalizability to unseen data.

Accurate segmentation of lesions is crucial for disease diagnosis and treatment planning. However, blurring and low contrast in the imaging process can affect segmentation results. We have observed that noninvasive medical imaging shares considerable similarities with natural images under low light conditions and that nocturnal animals possess extremely strong night vision capabilities. Inspired by the dark vision of these nocturnal animals, we proposed a novel plug-and-play dark vision network (DVNet) to enhance the model's perception for low-contrast medical images. Specifically, by employing the wavelet transform, we decompose medical images into subbands of varying frequencies, mimicking the sensitivity of photoreceptor cells to different light intensities. To simulate the antagonistic receptive fields of horizontal cells and bipolar cells, we design a Mamba-Enhanced Fusion Module to achieve global information correlation and enhance contrast between lesions and surrounding healthy tissues. Extensive experiments demonstrate that the DVNet achieves SOTA performance in various medical image segmentation tasks.

We propose AutoSense, a millimeter-wave (mmWave) wireless signal-based system for predicting 3D bounding boxes of vehicles. While cameras and LiDAR can be adversely affected by challenging weather conditions such as heavy rain, fog, or snow, mmWave signals are less susceptible to these environmental factors, making them more resilient. As a result, AutoSense can complement other sensors for accurate 3D bounding box predictions in all weather conditions.

Millimeter-wave is the core technology to enable multi-Gbps throughput and ultra-low latency connectivity. But the devices need to operate at very high frequency and ultra-wide bandwidth: They consume more energy, dissipate more power, and subsequently heat up faster. Device overheating is a common concern of many users, and millimeter-wave would exacerbate the problem. In this work, we first thermally characterize millimeter-wave devices. Our measurements reveal that after only 10 s of data transfer at 1.9 Gbps bit-rate, the millimeter-wave antenna temperature reaches 68 °C; it reduces the link throughput by 21%, increases the standard deviation of throughput by 6×, and takes 130 s to dissipate the heat completely. Besides degrading the user experience, exposure to high device temperature also creates discomfort. Based on the measurement insights, we propose Aquilo, a temperature-aware, multi-antenna network scheduler. It maintains relatively high throughput performance but cools down the devices substantially. Our testbed experiments under both static and mobile conditions demonstrate that Aquilo achieves a median peak temperature only 0.5 °C to 2 °C above the optimal while sacrificing less than 10% of throughput.

Soluble Sugar Content (SSC) of a fruit is indicative of its ripeness and is used in the fruit industry for quality control in the production chain. We present the design and implementation of SSCense, a low-cost, non-destructive system to estimate a fruit's SSC using the millimeter-wave wireless technology in 5G-and-beyond devices.

During convergence, after a link state change in traditional networks with a distributed control plane, packets may get caught in transient forwarding loops. Such loops can be avoided by imposing a certain order among the routers in updating their forwarding information bases (FIBs), but it requires some form of coordination among routers. As an alternative, a progressive link metric increment method has been proposed for loop-free forwarding without ordered FIB updates, but it takes longer to converge to the target state. In this paper, we show that the order of updates rarely matters for loop-free convergence when the failure inference-based fast reroute (FIFR) scheme with interface-specific forwarding is employed for dealing with link failures. The key insight is to have each router install the traditional interface-independent forwarding entries as soon as they are recomputed during convergence and install the recomputed interface-specific backwarding entries post-convergence. Our evaluation of 280 real and random topologies confirms that the order of updates does not matter with the proposed approach for 17336 out of 17339 links in those topologies. To handle such rare cases where the order matters, it can be coupled with progressive link metric increments to ensure loop-freedom with unordered FIB updates. Thus, the proposed approach, referred to as FIFR++, makes it possible to achieve disruption-free fast convergence and fast reroute without requiring any modification to the IP datagram and without needing any coordination between routers.

We present MilliCam, a system that captures the shape of small metallic objects, such as a gun, through obstructions, like clothing. MilliCam builds on the millimeter-wave (mmWave) imaging systems, which are widely used today in airport security checkpoints. Existing systems achieve high-resolution using a Synthetic Aperture Radar (SAR) principle, but require bulky motion controllers to position the mmWave device precisely. In contrast, MilliCam emulates the SAR principle by pure hand-swiping. However, alias-free, high-resolution imaging requires a linear, error-free hand-swiping motion. Furthermore, image focusing on an object of interest requires steering perfectly-shaped beam over the target-scene; but it is unavailable in off-the-shelf devices. We design a set of algorithms to enable high-quality handheld imaging: compensating for the errors in hand-swipe motion; and focusing the target-scene digitally without beam-steer. We have prototyped MilliCam on a 60 GHz testbed. Our experiments demonstrate that MilliCam can effectively combat motion errors and focus on the object in target-scene.

Millimeter-wave is the core technology to enable multi-Gbps throughput and ultra-low latency wireless connectivity. But the devices need to operate at very high frequency and ultra-wide bandwidth; so, they consume more energy, dissipate more power, and subsequently heat up faster. Device overheating is a common concern of many users, and millimeter-wave (mmWave) would exacerbate the problem. In this work, we first study the thermal characterization of mmWave devices. Our measurements reveal that after only 10 s. of data transfer at 1.9 Gbps bit-rate, the mmWave antenna temperature reaches 68°C; it reduces the link throughput by 21%, increases the standard deviation of throughput by 6×, and takes 130 s. to dissipate the heat completely. Besides degrading the user experience, exposure to high device temperature also creates discomfort. Based on the measurement insights, we propose Aquilo, a temperature-aware multi-antenna scheduler; it maintains relatively high throughput performance, but cools down the devices substantially. Our testbed experiments in both static and mobile conditions show that Aquilo reaches a median peak temperature just 0.5 to 2°C above the optimal by sacrificing less than 10% of throughput.

Millimeter-wave devices operate at very high frequency and ultra-wide bandwidth. They consume more energy, dissipate more power, and heat up faster. So, millimeter-wave (mmWave) would exacerbate the device overheating problem in the future. In this work, we first perform a thermal characterization of mmWave devices: it reveals that after only 10 s. of data transfer at 1.9 Gbps, the antenna temperature reaches 68°C; it reduces the link throughput by 21%, increases the standard deviation by 6×, and takes 130 s. to dissipate the heat completely. We then propose Aquilo to bring temperature-awareness in mmWave networks; Aquilo maintains relatively high throughput performance and cools down the devices substantially. Our testbed experiments in static conditions show that Aquilo reaches a median peak temperature just 1°C above the optimal with less than 10% throughput sacrifice only.

Breakthroughs in deep learning, GPUs, and edge computing have paved the way for always-on, live video analytics. However, to achieve real-time performance, a GPU needs to be dedicated amongst a few video feeds. But, GPUs are expensive resources and a large-scale deployment requires supporting hundreds of video cameras - exorbitant cost prohibits widespread adoption. To ease this burden, we propose Tetris, a system comprising of several optimization techniques from computer vision and deep-learning literature blended in a synergistic manner. Tetris is designed to maximize the parallel processing of video feeds on a single GPU, with a marginal drop in inference accuracy. Tetris performs CPU-based tiling of active regions to combine activities across video feeds. resulting in a condensed input volume. It then runs the deep learning model on this condensed volume instead of individual feeds, which significantly improves the GPU utilization. Our evaluation on Duke MTMC dataset reveals that Tetris can process 4x video feeds in parallel compared to any of the existing methods used in isolation.

This paper is focused on providing fast reroute and loop-free convergence in traditional IP networks, without making any modifications to the IP datagram and without requiring any coordination between routers for FIB updates. Failure inference based fast route (FIFR) is an approach in which routers adjacent to a failed link or router perform local rerouting around the failure, without notifying non-adjacent routers about the failure. The non-adjacent routers utilize interface-specific forwarding tables, which are precomputed based on potential inferred failures that could cause a packet for a given destination to arrive through that unusual interface, to ensure loop-free forwarding towards the destination. However, as long as the failure lasts, packets that were to be forwarded over the failed link traverse suboptimal paths, as they reach the router adjacent to the failure and then are rerouted along a detour. Therefore, in case of a long-lasting failure, it is desirable to trigger a network-wide link state update, so that all routers can converge to new optimal forwarding tables. But, without some coordination between routers to install their forwarding entries in a specific order, there may be transient forwarding loops during the convergence period. As we are interested in a mechanism that does not require any such coordination between routers, we consider the possibility of employing progressive link state updates. In this paper, we show that FIFR with progressive link metric increments can guarantee loop-free forwarding not only before/after but during convergence too and protect against non-partitioning single link failures.

This paper explores the possibility of injecting mobility into wireless network infrastructure. We envision WiFi APs on wheels that move to optimize user performance. Movements need not be all around the floor, neither do they have to operate on batteries. As a first step, WiFi APs at home could remain tethered to power and Ethernet outlets while moving in small areas (perhaps under the couch). If such systems prove successful, perhaps future buildings could offer explicit support for network infrastructure mobility. This paper begins with a higher level discussion of robotic wireless networks-the opportunities and the hurdles-and then pivots by developing a smaller slice of the vision through a system called iMob. With iMob, a WiFi AP is mounted on a Roomba robot and made to periodically move within a 2x2 sqft region. The core questions pertain to finding the best location to move to, such that the SNRs from its clients are strong, and the interferences from other APs are weak. Our measurements show that the richness of wireless multipath offers significant opportunities-even within a 2x2 sqft region, locations exist that are 1:7x better than the average location in terms of throughput. When multiple APs in a neighborhood coordinate, the gains can be even higher. In sum, although infrastructure mobility has been discussed in the context of Google Balloons, ad hoc networks, and delay tolerant networks, we believe that the possibility of moving our personal devices in homes and offices is relatively unexplored, and could open doors to new kinds of innovation.

This paper studies the feasibility of minimizing convergence delay and forwarding disruption without carrying any additional bits in the IP header, to provide high availability despite link failures in traditional IP networks. Previously proposed mechanisms achieve two of these three objectives by trading off the other objective. For instance, the ordered forwarding information base updates approach may prolong the convergence delay, whereas the SafeGuard scheme requires carrying the path cost in the IP header. As a better alternative, we propose a scheme called fast convergence with fast reroute (FCFR), which combines the features of IP fast rerouting and interface-specific forwarding. We show that FCFR can achieve minimal convergence delay, while ensuring loop-free delivery during convergence, after a single non-partitioning failure in an IP network, without altering the IP header format, making it amenable for immediate deployment.

Most location based services require semantic place names such as Staples, rather than physical coordinates. Past work has mostly focussed on achieving localization accuracy, while assuming that the translation of physical coordinates to semantic names will be done manually. This paper makes an effort to automate this step, by leveraging the presence of a website corresponding to each store and the availability of a repository of WiFi-tagged pictures from different stores. By correlating the text inside the pictures, against the text extracted from store websites, our proposed system, called AutoLabel, can automatically label clusters of pictures, and the corresponding WiFi APs, with store names. Later, when a user enters a store, her mobile device scans the WiFi APs and consults a lookup table to recognize the store she is in. Experiment results from 40 different stores show recognition accuracy upwards of 87%, even with as few as 10 pictures from a store, offering hope that automatic large-scale semantic labeling may indeed be possible from pictures and websites of stores.

Randomized backoff is a well-established approach for avoiding collisions in CSMA networks. Today's backoff operation, such as in WiFi, attempts to create a total ordering among all the nodes contending for the channel. Total ordering requires assigning a unique backoff to each node, which is achieved by having nodes choose their back-offs from a large range, ultimately leading to channel wastage. This paper observes that total ordering can be achieved more efficiently. We propose “hierarchical backoff” in which nodes pick random numbers from a smaller range, resulting in groups of nodes picking the same number (i.e., partial order). Now, the group of nodes that picks the smallest number is advanced to a second round, where they again perform the same operation. This results in more efficient backoff because the time for partially ordering all nodes plus totally ordering each small groups is actually less than the time needed to totally order all nodes. Realizing the above intuition requires addressing new protocol challenges in group signaling, the feasibility of which is demonstrated on a USRP/GNUradio prototype. Large scale simulations also show consistent throughput gains by incorporating the proposed backoff approach into two CSMA protocols - WiFi and oCSMA. We also show that the proposed approach can be complementary to and even outperform existing backoff optimization schemes.

Due to the popularity of soccer and the purposeful use of the head during play, traumatic brain injury to soccer players has been a concern for decades. However, there is a sense of urgency now in understanding and preventing concussions better, due to raising public awareness. Towards that end, intra-oral devices such as Vector MouthGuards are being studied [3] to measure the athlete's head's linear and rotational accelerations from impacts experienced in practices and games. But given the players' natural distaste for such intra-oral devices, more palatable alternatives for head impact monitoring are being developed [2]. X2 Biosystems xPatchis an electronic skin patch thatis worn behind the ear. Reebok Checklight embeds the impact sensor in the back of a skullcap which can be worn with or without a helmet. Triax SIM-P is placed inside a headband for non-helmeted sports and a skullcap for helmeted sports. While all these devices are much more convenient to wear than intra-oral devices, it is yet to be seen whether they gain wider acceptance, particularly by the millions of amateur soccer players all over the world.

This paper develops techniques using which humans can be visually recognized. While face recognition would be one approach to this problem, we believe that it may not be always possible to see a person?s face. Our technique is complementary to face recognition, and exploits the intuition that human motion patterns and clothing colors can together encode several bits of information. Treating this information as a "temporary fingerprint", it may be feasible to recognize an individual with reasonable consistency, while allowing her to turn off the fingerprint at will.

No abstract available.

Mobile phones are becoming the convergent platform for personal sensing, computing, and communication. This paper attempts to exploit this convergence toward the problem of automatic image tagging. We envision TagSense, a mobile phone-based collaborative system that senses the people, activity, and context in a picture, and merges them carefully to create tags on-the-fly. The main challenge pertains to discriminating phone users that are in the picture from those that are not. We deploy a prototype of TagSense on eight Android phones, and demonstrate its effectiveness through 200 pictures, taken in various social settings. While research in face recognition continues to improve image tagging, TagSense is an attempt to embrace additional dimensions of sensing toward this end goal. Performance comparison with Apple iPhoto and Google Picasa shows that such an out-of-band approach is valuable, especially with increasing device density and greater sophistication in sensing and learning algorithms.

Today's smartphones provide a variety of sensors, enabling high-resolution measurements of user behavior. We envision that many services can benefit from short-term predictions of complex human behavioral patterns. While enablement of behavior awareness through sensing is a broad research theme, one possibility is in predicting how quickly a person will move through a space. Such a prediction service could have numerous applications. For one example, we imagine shop owners predicting how long a particular customer is likely to browse merchandise, and issue targeted mobile coupons accordingly - customers in a hurry can be encouraged to stay and consider discounts. Within a space of moderate size, WiFi access points are uniquely positioned to track a statistical framework for user length of stay, passively recording metrics such as WiFI signal strength (RSSI) and potentially receiving client-uploaded sensor data. In this work, we attempt to quantity this opportunity, and show that human dwell time can be predicted with reasonable accuracy, even when restricted to passively observed WiFi RSSI.

Successive interference cancellation (SIC) is a PHY capability that allows a receiver to decode packets that arrive simultaneously. While the technique is well known in communications literature, emerging software radio platforms are making practical experimentation feasible. This motivates us to study the extent of throughput gains possible with SIC from a MAC layer perspective and scenarios where such gains are worth pursuing. We find that contrary to our initial expectation, the gains are not high when the bits of interfering signals are not known a priori to the receiver. Moreover, we observe that the scope for SIC gets squeezed by the advances in bitrate adaptation. In particular, our analysis shows that interfering one-to-one transmissions benefit less from SIC than scenarios with many-to-one transmissions (such as when clients upload data to a common access point). In view of this, we develop an SIC-aware scheduling algorithm that employs client pairing and power reduction to extract the most gains from SIC. We believe that our findings will be useful guidelines for moving forward with SIC-aware protocol research.

Kinect has become quite popular for gaming as it tracks players' natural gestures without a controller like other gaming systems. But it has some inherent limitations such as occlusion problem and fails to track a player accurately if there is an obstacle. In this work, we propose to leverage smartphone to supplement Kinect. We explore how to fuse information from Kinect's tracking with the smartphone's sensor readings to improve Kinect gaming experience.

Motor vehicle accidents are one of the leading causes of death. While lane departure warning, blind spot warning, and driver attention monitoring systems for avoiding collisions have been in development for quite sometime, to date mostly luxury cars are only equipped with these safety features. As a cheaper and ubiquitous alternative, we explore how a smartphone can assist an inattentive driver by leveraging its front and back cameras apart from other sensors. The challenge, however, is given the resource constraints of a smartphone, how quickly and accurately can it detect an unintended maneuver and alert the driver. In this paper, we describe our on-going attempt to address this challenge.

It has been observed that transient failures are fairly common in IP backbone networks and there have been several proposals based on local rerouting to provide high network availability despite failures. While most of these proposals are effective in handling single failures, they either cause loops or drop packets in the case of multiple independent failures. To ensure forwarding continuity even with multiple failures, we propose Localized On-demand Link State (LOLS) routing. Under LOLS, each packet carries a blacklist, which is a minimal set of failed links encountered along its path, and the next hop is determined by excluding the blacklisted links. We show that the blacklist can be reset when the packet makes forward progress towards the destination and hence can be encoded in a few bits. Furthermore, blacklist-based forwarding entries at a router can be precomputed for a given set of failures requiring protection. While the LOLS approach is generic, this paper describes how it can be applied to ensure forwarding to all reachable destinations in case of any two link or node failures. Our evaluation of this failure scenario based on various real network topologies reveals that LOLS needs 6 bits in the worst case to convey the blacklist information. We argue that this overhead is acceptable considering that LOLS routing deviates from the optimal path by a small stretch only while routing around failures.

Modern wireless interfaces support a physical-layer capability called Message in Message (MIM). Briefly, MIM allows a receiver to disengage from an ongoing reception and engage onto a stronger incoming signal. Links that otherwise conflict with each other can be made concurrent with MIM. However, the concurrency is not immediate and can be achieved only if conflicting links begin transmission in a specific order. The importance of link order is new in wireless research, motivating MIM-aware revisions to link-scheduling protocols. This paper identifies the opportunity in MIM-aware reordering, characterizes the optimal improvement in throughput, and designs a link-layer protocol for enterprise wireless LANs to achieve it. Testbed and simulation results confirm the performance gains of the proposed system.

A wireless transmitter learns of a packet loss and infers collision only after completing the entire transmission. If the transmitter could detect the collision early [such as with carrier sense multiple access with collision detection (CSMA/CD) in wired networks], it could immediately abort its transmission, freeing the channel for useful communication. There are two main hurdles to realize CSMA/CD in wireless networks. First, a wireless transmitter cannot simultaneously transmit and listen for a collision. Second, any channel activity around the transmitter may not be an indicator of collision at the receiver. This paper attempts to approximate CSMA/CD in wireless networks with a novel scheme called CSMA/CN (collision notification). Under CSMA/CN, the receiver uses PHY-layer information to detect a collision and immediately notifies the transmitter. The collision notification consists of a unique signature, sent on the same channel as the data. The transmitter employs a listener antenna and performs signature correlation to discern this notification. Once discerned, the transmitter immediately aborts the transmission. We show that the notification signature can be reliably detected at the listener antenna, even in the presence of a strong self-interference from the transmit antenna. A prototype testbed of 10 USRP/GNU Radios demonstrates the feasibility and effectiveness of CSMA/CN.

Conventional WiFi networks perform channel contention in time domain. This is known to be wasteful because the channel is forced to remain idle while all contending nodes are backing off for multiple time slots. This paper proposes to break away from convention and recreate the backing off operation in the frequency domain. Our basic idea leverages the observation that OFDM subcarriers can be treated as integer numbers. Thus, instead of picking a random backoff duration in time, a contending node can signal on a randomly chosen subcarrier. By employing a second antenna to listen to all the subcarriers, each node can determine whether its chosen integer (or subcarrier) is the smallest among all others. In fact, each node can even determine the rank of its chosen subcarrier, enabling the feasibility of scheduled transmissions after every round of contention. We develop these ideas into a Back2F protocol that migrates WiFi backoff to the frequency domain. Experiments on a prototype of 10 USRPs confirm feasibility, along with consistent throughput gains over 802.11. at high bit rates. Trace based simulations affirm scalability to larger, real-world network topologies.

Mobile phones are becoming the convergent platform for personal sensing, computing, and communication. This paper attempts to exploit this convergence towards the problem of automatic image tagging. We envision TagSense, a mobile phone based collaborative system that senses the people, activity, and context in a picture, and merges them carefully to create tags on-the-fly. The main challenge pertains to discriminating phone users that are in the picture from those that are not. We deploy a prototype of TagSense on 8 Android phones, and demonstrate its effectiveness through 200 pictures, taken in various social settings. While research in face recognition continues to improve image tagging, TagSense is an attempt to embrace additional dimensions of sensing towards this end goal. Performance comparison with Apple iPhoto and Google Picasa shows that such an out-of-band approach is valuable, especially with increasing device density and greater sophistication in sensing/learning algorithms.

Smartphones are becoming the convergent platform for personal sensing, computing, and communication. Our work attempts to exploit this convergence towards the problem of automatic image tagging. We envision TagSense, a smartphone based collaborative system that senses the people/activity/context in a picture, and merges them carefully to create tags on-the-fly. The main challenge pertains to discriminating phone users that are in the picture, from those that are not. Our demonstration system consists of 8 Android phones and a laptop. Phones -- with the TagSense application running -- will be randomly distributed to participants. Once a picture is taken by a participant with the phone, tags generated by TagSense will be shown on the phone screen.

Conventional WiFi networks perform channel contention in time domain. This is known to be wasteful because the channel is forced to remain idle, while all contending nodes are backing off for multiple time slots. This paper proposes to break away from convention and recreate the backing off operation in the frequency domain. Our basic idea is to pretend that OFDM subcarriers are integer numbers, and thereby, view today's random backoff process as equivalent to transmitting on a randomly chosen subcarrier. By employing a second antenna to listen to all the subcarriers, each node can determine whether its chosen integer (or subcarrier) is the smallest among all others. In fact, each node can even determine the rank of its chosen integer, enabling the feasibility of a TDMA-like schedule from every round of contention. We develop these ideas into a Time to Frequency (T2F) protocol and prototype it on a small testbed of 8 USRPs. Experiments confirm its feasibility, along with promising throughput gains of more than 35% at high bit rates. A fuller design and thorough evaluation of T2F is a topic of ongoing work.

Successive interference cancellation (SIC) is a physical layer capability that allows a receiver to decode packets that arrive simultaneously. While the technique is well known in communications literature, emerging software radios are making practical experimentation feasible. This motivates us to study the extent of throughput gains possible with SIC from a MAC layer perspective. Contrary to our initial expectation, we find that the gains from SIC are not easily available in many realistic situations. Moreover, we observe that the scope for SIC gets squeezed by the advances in bitrate adaptation, casting doubt on the future of SIC based protocols.

It has been observed that even in well managed networks, failures of links and routers are not uncommon. In order to satisfy the demand for high availability in case of a failure, fast restoration of loop-free forwarding along the optimal paths is imperative for a routing scheme. Failures can be circumvented quickly with local rerouting but packets take potentially long detours. Global recomputation of new optimal routes incurs a convergence delay and can cause forwarding loops during convergence. Attempts to avoid transient loops may also increase the convergence delay. The recently proposed SafeGuard mechanism overcomes these problems, i.e., it is always loop-free, and minimizes disruption time and convergence delay. One drawback however, is that SafeGuard needs each packet to carry multiple-byte information about the path cost. We propose an alternative approach, fast convergence with fast reroute (FCFR), that employs a fast reroute scheme such as NotVia and needs just one additional bit in the packet header. We evaluate the performance of FCFR, and show that it performs comparably to SafeGuard, with much less per-packet overhead.

The nature of human mobility demands that mobile devices become agile to diverse operating environments. Coping with such diversity requires the device to assess its environment, and trigger appropriate responses to each of them. While existing communication subsystems rely on in-band wireless signals for context-assessment and response, we explore a lateral approach of using out-of-band sensor information. We propose a relatively novel framework that synthesizes in-band and out-of-band information, facilitating more informed communication decisions. We believe that further research in this direction could enable a new kind of device agility, deficient in today's communication systems. Since such a framework is located at the boundaries of mobile sensing and wireless communication, we call it sensor assisted wireless communication.

A wireless transmitter learns of a packet loss, infers collision, only after completing the entire transmission. If the transmitter could detect the collision early (such as with CSMA/CD in wired networks), it could immediately abort its transmission, freeing the channel for useful communication. There are two main hurdles to realize CSMA/CD in wireless networks. First, a wireless transmitter cannot simultaneously transmit and listen for a collision. Second, any channel activity around the transmitter may not be an indicator of collision at the receiver. This paper attempts to approximate CSMA/CD in wireless networks with a scheme called CSMA/CN (collision notification). Under CSMA/CN, the receiver uses PHY layer information to detect a collision and immediately notifies the transmitter. The collision notification consists of a unique signature, sent on the same channel as the data. The transmitter employs a listener antenna and performs signature correlation to discern this notification. Once discerned, the transmitter immediately aborts transmission. We show that the notification signature can be reliably detected at the listener antenna, even in the presence of a strong self-interference from the transmit antenna. A prototype testbed of 10 USRP/GNURadios demonstrates the feasibility and effectiveness of CSMA/CN

Modern wireless interfaces support a physical layer capability called Message in Message (MIM). Briefly, MIM allows a receiver to disengage from an ongoing reception, and engage onto a stronger incoming signal. Links that otherwise conflict with each other, can be made concurrent with MIM. However, the concurrency is not immediate, and can be achieved only if conflicting links begin transmission in a specific order. The importance of link order is new in wireless research, motivating MIM-aware revisions to link scheduling protocols. This paper identifies the opportunity in MIM-aware reordering, characterizes the optimal improvement in throughput, and designs a link layer protocol to achieve it. Testbed results confirm the performance gains of the proposed system.

Neighbors of both the transmitter and the receiver must keep quiet in a 802.11 wireless network as it requires bidirectional exchange, i.e., nodes reverse their roles as transmitters and receivers, for transmitting a single DATA frame. To reduce role reversals and to improve spatial reuse, a piggybacked acknowledgment based approach has been proposed to enable concurrent transmissions. Recent findings on physical layer capture show that it is possible to capture a frame of interest in the presence of concurrent interference and that the SINR threshold is dependent on the relative order in which the frame and the interference arrive at the receiver. In this paper, we show that it is possible to exploit capture and increase concurrent transmissions in wireless adhoc networks. We develop a distributed channel access scheme and demonstrate that it offers significant throughput gain particularly at lower data rates.

802.11 requires bidirectional exchange (i.e., nodes reverse their roles as transmitters and receivers) and hence neighbors of both the transmitter and receiver must keep quiet for the entire duration of communication. This degrades spatial reuse, leading to low network throughput. To address this problem, power control, rate control, and carrier-sense adaptations have successfully identified possibilities of concurrency in the spatial domain. In the temporal domain, optimizations such as piggybacked Acks [1] have reduced role reversals, also enabling concurrency. Though beneficial, these improvements are bounded by the SINR requirement. Recent studies found that the SINR threshold is a dynamic value, dependent on the relative order in which the signal and the interference arrive at the receiver. This implies that under certain conditions, it might be feasible to capture a data frame in the presence of concurrent interference. If harnessed carefully, this can help improve the spatial reuse of wireless networks.

It has been observed that transient failures are fairly common in IP backbone networks and there have been several proposals based on local rerouting to provide high network availability despite transient failures. Previously, we proposed failure inferencing based fast rerouting for IP backbone networks that ensures delivery of a packet to its destination if there exists a path when a single link fails but can cause forwarding loops in case of multiple simultaneous failures. On the other hand, blacklist-aided forwarding, we proposed earlier for wireless mesh networks, provides loop-free forwarding even in the presence of multiple failed links in the network but requires that each packet carry a blacklist of failed links encountered along its path. Our aim is to achieve the best of both these approaches, i.e., successfully deliver packets while ensuring loop-freedom even in case of multiple failures without changing packet format. We propose blacklist-based interface-specific forwarding (BISF) that infers a blacklist, a list of links that might have failed, based on a packet's incoming interface and its destination, and determines the next-hop by excluding the blacklisted links. We show that BISF is loop-free regardless of the number of failures in the network while forwarding packets successfully in most cases.

With the emergence of voice over IP and other real-time business applications, there is a growing demand for an IP network with high service availability. Unfortunately, in today's Internet, transient failures occur frequently due to faulty interfaces, router crashes, etc., and current IP networks lack the resiliency needed to provide high availability. To enhance availability, we proposed failure inferencing based fast rerouting (FIFR) approach that exploits the existence of a forwarding table per line-card, for lookup efficiency in current routers, to provide fast rerouting similar to MPLS, while adhering to the destination-based forwarding paradigm. In our previous work, we have shown that the FIFR approach can deal with single link failures. In this paper, we extend the FIFR approach to ensure loop-free packet delivery in case of single router failures also, thus mitigating the impact of many scenarios of failures. We demonstrate that the proposed approach not only provides high service availability but also incurs minimal routing overhead.

Abstract — Static broadband wireless networks, due to their ease of deployment, are likely to proliferate in the near future. The major stumbling block, however, is that wireless links are prone to external interference, channel fading, inclement weather, etc. Therefore scalable and reliable routing despite frequent link quality fluctuations is needed for accelerating the growth of these networks. Most of the wireless routing schemes proposed in the literature are less suitable for these networks, as they are designed primarily for mobile ad hoc networks with dynamic and unpredictable topologies. In this paper, we propose a novel link-state-based blacklist-aided forwarding (BAF) approach, that takes advantage of the fact that the nodes and therefore their adjacencies are relatively static, for scalable packet delivery in static wireless networks. Under BAF, each packet carries a blacklist, a minimal set of degraded-quality links encountered along its path, and the next hop is determined based on both its destination and blacklist. BAF provides loop-free delivery of packets to reachable destinations regardless of the number of degraded links in the network. We evaluate the performance of BAF and show that it is not only reliable but also scalable. I.

Multi-homing is a common practice among many (especially large) customer (or stub) networks. Although the purpose of multi-homing is primarily for enhanced reliability, it has also increasingly been used for load balancing and latency reduction. In this paper, we address the problem of how to perform scalable route selection in a multi-homed stub network to optimize network latency to various destinations as measured by round-trip-time (RTT). A straight forward method is to simply perform RTT measurements (e.g., using ping) to each destination via each provider and select the one with the minimum RTT as the "best" next-hop to the destination. Is there a more.

Dealing with network failures effectively is a major operational challenge for Internet service providers. Commonly deployed link state routing protocols such as OSPF react to link failures through global (i.e., network-wide) link state advertisements and routing table recomputations, causing significant forwarding discontinuity after a failure. The drawback with these protocols is that they need to trade off routing stability and forwarding continuity. To improve failure resiliency without jeopardizing routing stability, we propose a proactive local rerouting based approach called failure insensitive routing (FIR). The proposed approach prepares for failures using interface-specific forwarding, and upon a failure, suppresses the link state advertisement and instead triggers local rerouting using a backwarding table. In this paper, we prove that when no more than one link failure notification is suppressed, FIR always finds a loop-free path to a destination if one such path exists. We also formally analyze routing stability and network availability under both proactive and reactive approaches, and show that FIR provides better stability and availability than OSPF.

Communication between a pair of nodes in the network may get disrupted due to failures of links/nodes resulting in zero effective bandwidth between them during the recovery period. It has been observed that such disruptions are not too uncommon and may last from tens of seconds to minutes. Even an occasional such disruption can drastically degrade the viewing experience of a participant in a video streaming session particularly when a sequence of frames central to the story are lost during the disruption. The conventional approach of prefetching video frames and patching lost ones with retransmissions is not always viable when disruptions are localized and experienced only by a few among many receivers. Error spreading approaches that distribute the losses across the video work well only when the disruptions are quite short. As a better alternative, we propose a disruption-tolerant content-aware video streaming approach that combines the techniques of content summarization and error spreading to enhance viewers experience even when the disruptions are long. We introduce the notion of "substitutable content summary frames" and provide a method to select these frames and also their transmission order to mitigate the impact of a disruption. In the event of a disruption, the already received summary frames are played by the client during disruption and near normal playback is resumed after the disruption. We evaluate our approach and demonstrate that it provides acceptable viewing experience with minimal startup latency and client buffer.

Most of the QoS routing schemes proposed so far require periodic exchange of QoS state information among routers, imposing both communication overhead on the network and processing overhead on core routers. Furthermore, stale QoS state information causes the performance of these QoS routing schemes to degrade drastically. In order to circumvent these problems, we focus on localized QoS routing schemes where the edge routers make routing decisions using only local information and thus reducing the overhead at core routers. We first describe virtual capacity based routing (vcr), a theoretical scheme based on the notion of virtual capacity of a route. We then propose proportional sticky routing, an easily realizable approximation of vcr and analyze its performance. We demonstrate through extensive simulations that adaptive proportional routing is indeed a viable alternative to the global QoS routing approach.

Most of the QoS routing schemes proposed so far require periodic exchange of QoS state information among routers, imposing both communication overhead on the network and processing overhead on core routers. Furthermore, stale QoS state information causes the performance of these QoS routing schemes to degrade drastically. In order to circumvent these problems, we focus on localized QoS routing schemes where the edge routers make routing decisions using only "local" information and thus reducing the overhead at core routers. We first describe virtual capacity-based routing (VCR), a theoretical scheme based on the notion of virtual capacity of a route. We then propose proportional sticky routing (PSR), an easily realizable approximation of VCR and analyze its performance. We demonstrate through extensive simulations that adaptive proportional routing is indeed a viable alternative to the global QoS routing approach.

Video delivery from a server to a client across a network is an important component of many multimedia applications. While delivering a video stream across a resource constrained network, loss of frames may be unavoidable. Under such circumstances, it is desirable to find a server transmission schedule that can efficiently utilize the network resources while maximizing the perceived quality-of-service (QoS) at the client. To address this issue, we introduce the notion of selective frame discard at the server and formulate the optimal selective frame discard problem using a QoS based cost function. Given network bandwidth and client buffer constraints, we develop an O(N log N) algorithm to find the minimum number of frames that must be discarded in order to meet these constraints. The correctness of the algorithm is also formally established. Since the computational complexity of the optimal algorithm for solving the optimal selective frame discard problem is prohibitively high in general, we also develop several efficient heuristic algorithms for selective frame discard. These algorithms are evaluated using JPEG video traces.