Mission-critical applications in future vehicular networks require highly reliable wireless communications. Multi-connectivity is a potential solution to meet the desired quality-of-services of these applications. In this work, we consider a highway scenario with multi-connectivity where a vehicle can combine packets transmitted by multiple remote-radio-heads. For given inter-site- distances, number of links, and frequency reuse factors, we derive the expression for outage probability considering communication links affected by shadowing. We investigate the effect of network deployment parameters on the achievable outage reliability under different radio parameters. The simulation results show trade-offs between inter-site distance and control parameters (frequency reuse factor and number of links), as well as network and radio channel parameters (number of satisfied users, effect of shadowing, and path loss exponent) on communications reliability.

Emerging wireless communication use-cases demand ultra-reliable and low-latency communications. In order to meet these requirements, multi-connectivity (MC) approaches are being considered as a possible solution. To enable multi-connectivity with efficient use of resources, link adaptation requires not only to adopt modulation and coding scheme but also the number of links. This can be achieved by using an effective link quality metric (LQM) and mapping it to packet error rate or throughput, a process known as physical layer abstraction (PLA). In this paper, a new PLA method enhanced received bit information rate (eRBIR) is presented for OFDM based MC networks. The performance of proposed method is evaluated and compared against existing methods such as exponential effective SINR mapping (EESM) and received bit information rate (RBIR). Simulation results show that the proposed method enable an accurate and reliable link adaptation as compared to state of the art PLAs. Finally, the application of PLAs, to adapt the MCS in varying channel conditions is illustrated to ensure a certain target quality of service.

Locating active users/transmitters and predicting interference patterns in real-time is a key challenge in next generation radio networks to enable robust wireless communication and highly dynamic radio resource allocation. We consider a network of collaborating spectrum sensing units (SUs) and propose a two-phase approach for localization and tracking of transmitters using received signal strength at SUs. The first phase involves estimation of locations of multiple transmitters using compressed sensing based method. In the second phase, the location estimates are improved and velocities of mobile transmitters are estimated using extended Kalman filter. The proposed approach is evaluated through simulations considering an industrial channel model with correlated shadowing. We investigate the effect of SU placement, transmitter density, mobility and fading on the localization performance.

The fifth generation (5G) of mobile communication will enable new use-cases such as self driving cars, smart automation and mission critical applications, which require ultra-reliable communications. Multi-Connectivity (MC) is a promising approach to achieve high reliability in wireless networks. In order to enable MC and efficiently utilize radio resources, dynamic link adaptation is required for choosing appropriate modulation schemes and number of links. This can be achieved by using an effective link quality metric such as effective signal-to-noise ratio (SNR) and mapping it to the packet error rate, a process referred to as physical layer abstraction (PLA). In this paper, we investigate and compare the performance of existing PLA methods especially exponential effective SNR mapping (EESM) and received bit information rate (RBIR), for OFDM-based MC systems. Furthermore, we propose a new robust PLA method, called enhanced EESM (eEESM). The eEESM minimizes the efforts of optimizing tuning parameter by fitting the variations in tuning parameter to known curves as a function of channel and diversity order. Simulation results show that eEESM outperforms the state of the art PLAs for different channel conditions, modulation and diversity orders.

HPC interconnect is a very crucial component of any HPC machine. Interconnect performance is one of the contributing factors for overall performance of HPC system. Most popular interface to connect Network Interface Card (NIC) to CPU is PCI express (PCIe). With denser core counts in compute servers and increasingly maturing fabric interconnect speeds, there is need to maximize the packet data movement throughput between system memory and fabric interface network buffers, such that the rate at which applications (CPU) generating data is matched with rate at which fabric consumes the same. Thus PCIe throughput for small and medium messages (Programmed Input/Output) needs to be improved, to synchronize with core processing rate and fabric speeds. And there is scope for this improvement by increasing the payload size in Transaction Layer Packet (TLP) of PCIe. Traditionally, CPU issues memory writes in 8 bytes (payload for TLP), underutilizing the PCIe bus since overhead compared to payload is more. Write combining can increase the payload size in TLP, leading to more efficient utilization of available bus bandwidth, thereby improving the overall throughput. This work evaluates the performance that could be gained by using Write Combine Buffers (WCB) available on Intel CPU, for send side interface of HPC interconnect. These buffers are used for aggregating the small (usually 8 bytes) memory mapped I/O stores, to form the bigger PCIe Transaction Level Packets (TLP), which leads to better bus bandwidth utilization. It is observed that, this technique improves peak PIO bandwidth by 2x compared to normal PIO. It is also observed that till 4096 bytes write combine enabled PIO outperforms DMA.

In this paper, we study cooperative cognitive radio networks consisting of a primary user (PU) and multiple secondary users (SUs). SUs transmit only when PU is sensed as silent and may interfere with primary transmission due to imperfect sensing. When primary activity is sensed correctly, SUs cooperate with PU by assisting retransmission of failed packets of PU. We analyze packet throughput of PU and SU for three variations of the proposed cooperation method. A signal flow graph-based approach is employed to obtain the closed-form expressions of packet throughput. The analysis is done for two cases: individual sensing and cooperative sensing. Furthermore, we characterize the optimal transmission probability of SUs that maximizes individual secondary packet throughput keeping all queues in the system stable. Results present a comparison of throughput performance of the proposed cooperation methods under different scenarios and show their benefits for both PU and SU throughput.

In this paper, we consider a cognitive radio system with a primary user and a secondary user, both powered by batteries that can recharge over time by harvesting energy from renewable sources. Both users store packets in queues for transmission. We consider a cooperation scenario where secondary user receives unsuccessful packets transmitted by primary user and relays them to primary receiver. For the system to be stable, all queues in the system should be stable. We find the stable throughput region for given system model. In battery powered transmitters, amount of energy available at a transmitter affects the activity of users, therefore affecting the stable throughput region of system. Based on the analysis of stability region, we present numerical results and study the effect of energy arrival rate as well as battery capacity on stability region and on delay of primary user packets. Finally, we compare the cooperation scheme with non-cooperation case for different energy arrival rates.

We consider a cognitive radio system in which primary user stores packets in a queue for transmission. For proper functioning of primary system, the cognitive user must ensure that primary queue remains stable. We show that such a queue stability constraint can be converted to an equivalent interference constraint on cognitive user's transmission. Under this constraint, we propose a power allocation technique to maximize sum of individual outage capacities achieved on each sub-channel of an OFDM based cognitive user. We present simulation results depicting effect of channel gains of direct links and interference links on primary user's stable throughput and cognitive user's sum-outage capacity. The results show that proposed method performs better than only interweave mode of transmission used in literature.

For cluster based parallel computing, optimization at every layer of Network Protocol to improve the performance is an ongoing process. Use of dedicated Network Processor has become good choice to cope up with continuous and rapid growth of network link speed. GPP or RISC can be used for packet processing, because of their higher processing speed and flexibility. To improve overall performance, it is necessary to program Network Processor by making efficient use of its resources. In our experiment, we have formed Hot Connection pool that is heavily dependent on Network Processor cache. Our experimental result shows performance improvement when pool value is chosen proportionate to cache size of Network Processor.