Personalized Federated Learning (PFL) enhances traditional Federated Learning (FL) by addressing the challenges associated with non-independent and identically distributed (non-IID) data and retains FL’s privacy-preserving feature. However, this decentralized architecture still presents security challenges. This study focuses on data poisoning attacks, particularly label flipping attacks in PFL, where some malicious clients upload poisoned model parameters to the server, which could then compromise the personalized models of honest clients during the server’s aggregation process. Moreover, we discovered that even with a small number of attackers, targeted data poisoning attacks are feasible and can significantly affect the model’s performance on specific labels. Finally, we propose four defensive strategies that have proven effective in identifying malicious clients. Additionally, we provide a framework that assists users in selecting the most suitable defense strategy based on their specific circumstances, thereby enhancing the robustness of the PFL architecture.

Motor Imagery (MI) is essential in the Brain-Computer Interface (BCI). Given the multi-source nature of EEG instability across subjects/sessions and the privacy concerns when dealing with data from other institutions, the importance of multi-source-free domain adaptation (MSFDA) becomes evident in conducting subject-independent applications. However, current MI approaches are lack of MSFDA for considering different sources’ importance, which may lead to negative transfer. In this work, we propose a novel two-stage MSFDA framework, namely Privacy-preserving Attention-Weighted domain adaptation (PAW), which is the first to emphasize different source importance in EEG MI by identifying crucial features across subjects and sessions. A domain discriminator is designed to find robust features within sessions in the training phase. In the adaptation phase, an attention weighted module determines the source importance, and a balanced confident set policy mitigates noise and imbalance samples. PAW delivers excellent performance via empirical evaluations on several public datasets. In particular, its adaptation phase is applicable across MSFDA scenarios, highlighting its broad application values.

The Long Short-Term Memory (LSTM) model has significantly improved time series prediction accuracy, but also brought forth concerns regarding reliability and security with its widespread adoption, particularly in the context of poisoning attacks. While there is substantial research on attacks and defenses for LSTM models, there’s limited focus on LSTM time series prediction models. In this paper, we propose an arithmetic-based poisoning attack methodology for a demonstrative LSTM time series speed prediction model. Furthermore, we employ the “red team/blue team exercises” commonly used in network security to develop defense strategies using support vector machine and linear regression analysis methods. Through the system-level simulation experiments, we verify the effectiveness of our proposed methodology. Our experiment results indicate that, regarding attacks, our methodology can identify the optimal attacks for the representative road segments. As for defenses, we demonstrate that the defended model’s performance is close to the real model’s performance.

Sensor failure detection and calibration is an important issue for Internet of Things applications. Many smart applications fail due to inaccurate data produced by their sensors. To solve this issue, we propose the CalibrationTalk mechanism. We use smart agriculture as an example to show how CalibrationTalk automatically fixes the sensing accuracy problems in commercial farm operations. Besides manufacturing variation, hardware malfunction and aging, sensing inaccuracy may be caused by soil erosion of rain or irrigation, which changes the positions and the angles of the sensors inserted in the soil. The above sensor failure problems can be detected by CalibrationTalk. In this article, we first extend our previous work to detect sensor failures. If the sensors become inaccurate due to aging or similar reasons, CalibrationTalk automatically calibrates the sensors in the farm fields. We conduct measurements and analytic modeling to investigate the performance of CalibrationTalk, which suggests that to detect a potential moisture sensor failure, it is appropriate to set the detection time as 30 min for moisture sensors. With this setup, true failure is detected while false detections rarely occur. For a detected sensor failure due to aging, CalibrationTalk can automatically build a calibration table. During the calibration table establishment period, the sensors still produce correct measured data and do not interfere with the normal operation of smart farming. The value of CalibrationTalk lies in its power of automation for sensor failure detection and calibration processes developed in this article.

In the broadband era, narrowband short message service (SMS) is still the most popular wireless data service. Many studies have been conducted to investigate the performance of SMS based on the arrival rates of short messages. From Chunghwa Telecom's commercial SMS call data records, we observed that even if the SMS arrival rates are the same, the distributions for the number of SMS arrivals per half hour are quite different for various observed days. We further identify that for the SMS traffic in a specific day, there are non-burst and burst periods. This paper investigates the SMS behaviors on weekdays, weekends, and holidays (specifically, new years' days and eves). With the assistance of kernel-based fitting method, we derive the SMS arrival number distributions of various traffic types and observed days. Our approach fits each SMS arrival number distribution by three cubic polynomial functions that can accurately capture the SMS behaviors. On the basis of the SMS arrival number distributions derived from our model, the mobile operators have better understanding about the volumes of short messages in different times and days, which can be used to design more flexible short message charging rates. Copyright © 2012 John Wiley & Sons, Ltd.

Vehicle speeds are often measured by intelligent transportation systems (ITSs) by utilizing sensors or software solutions. Our previous work proposed the Lin-Chang-Huangfu (LCH) scheme to compute the cell residence times by the standard counter values in mobile telecommunications switches. In this paper, we use mathematical and statistical developments to investigate the accuracy of the LCH scheme by deriving the bias of the cell residence times computed in this scheme. Then, we extend the LCH scheme with some filtering and compensation techniques for vehicle speed estimation and validate our approach with vehicle detector (VD) measurements at National Highway 3, Longtan Township, Taoyuan County, Taiwan. Our study indicates that the LCH scheme is an effective approach to vehicle speed estimation.

ZigBee, a unique communication standard designed for low-rate wireless personal area networks, has extremely low complexity, cost, and power consumption for wireless connectivity of inexpensive, portable, and moving devices. ZigBee uses the IEEE 802.15.4 standard as its communication protocol for medium access control (MAC) layer and physical (PHY) layer. The IEEE 802.15.4 MAC layer achieves duty-cycle operations by setting two system parameters, macBeaconOrder (BO) and macSuperFrameOrder (SO), to achieve low power consumption for ZigBee devices. This study comprehensively analyzes IEEE 802.15.4 duty-cycle operation. Specifically, a novel analytical model that accommodates a general traffic distribution is developed. An NS-2 based simulation model, which is validated by the developed analytical model is also proposed. Through the experiments conducted by the analytical and simulation models, some important performance-evaluation insights are gained that can be used as guidelines for future low-power ZigBee network deployment.

IEEE 802.15.4 is a new standard uniquely designed for low-rate wireless personal area networks. It targets ultralow complexity, cost, and power for low-rate wireless connectivity among inexpensive, portable, and moving devices. IEEE 802.15.4 provides a guaranteed time slot (GTS) mechanism to allocate a specific duration within a superframe for time-critical transmissions. This paper proposes an adaptive GTS allocation (AGA) scheme for IEEE 802.15.4, which considers low latency and fairness. The scheme is designed based on the existing IEEE 802.15.4 medium access control protocol, and IEEE 802.15.4 devices can receive this AGA service without any modification. A simulation model and an analytical model are developed to investigate the performance of our AGA scheme. The numerical results show that the proposed scheme significantly outperforms the existing IEEE 802.15.4 implementation.

The universal mobile telecommunications system (UMTS) utilizes the discontinuous reception (DRX) mechanism to reduce the power consumption of mobile stations (MSs). DRX permits an idle MS to power off the radio receiver for a predefined sleep period, and then wake up to receive the next paging message. The sleep/wake-up scheduling of each MS is determined by two DRX parameters: the inactivity timer threshold and the DRX cycle. In the literature, analytic and simulation models have been developed to study the DRX performance mainly for Poisson traffic. In this paper, we propose a novel semi-Markov process to model the UMTS DRX with bursty packet data traffic. The analytic results are validated against simulation experiments. We investigate the effects of the two DRX parameters on output measures including the power saving factor and the mean packet waiting time. Our study provides inactivity timer and DRX cycle value selection guidelines for various packet traffic patterns.

The IP multimedia core network subsystem (IMS) provides real-time multimedia services for Universal Mobile Telecommunications System (UMTS). Through recharge threshold-based credit reservation (RTCR) mechanism, prepaid IMS services can be supported by the online charging system (OCS) in UMTS. In RTCR, when the remaining amount of prepaid credit is below a threshold, the OCS reminds the user to recharge the prepaid account. It is essential to choose an appropriate recharge threshold to reduce the probability that the in-progress service sessions are forced-terminated. An analytic model is developed to investigate the performance of RTCR for the OCS. Based on our study, the network operator can select the appropriate parameter values for various traffic conditions.

In Universal Mobile Telecommunications System (UMTS), the extension of GPRS tunneling protocol called GTP' is utilized to transfer the charging data records (CDRs) from GPRS support nodes (GSNs) to charging gateways (CGs). To ensure that the mobile operator receives the charging information, availability for the GTP' transmission is essential. One important issue on GTP' availability is connection failure detection. It is desirable to select appropriate parameter values to avoid false failure detections (e.g., temporary network congestions) while to detect the true failures quickly. We propose an analytic model to compute the false failure detection probability and the expected true failure detection time. Based on our study, the network operator can select the appropriate parameter values for various traffic conditions to reduce the probability of false failure detection and/or true failure detection time.

The session timer mechanism for the session initiation protocol (SIP) has been proposed to track the states of the communicating sessions for proxy servers. Upon the occurrence of the session failure (e.g., radio link disconnection), the proxy server can quickly release the resources allocated for the failed session by using the SIP session timer. Based on the SIP session timer, we propose a dynamic session refreshing approach to adjust the session timer depending on the radio link conditions for wireless VoIP subscribers. The study indicates that with our dynamic session refreshing approach, the session failure can be efficiently detected without a considerable increase of signaling traffic.

In a cellular telecommunications network, the call blocking, forced termination, and call incompletion probabilities are major output measures of system performance. Most previous analytic studies assumed that the handover traffic to a cell is a fixed-rate Poisson process. Such assumption may cause significant inaccuracy in modeling. This paper shows that the handover traffic to a cell depends on the workloads of the neighboring cells. Based on this observation, we derive the exact equation for the handover force-termination probability when the mobile station (MS) cell residence times are exponentially distributed. Then, we propose an approximate model with general MS cell residence time distributions. The results are compared with a previously proposed model. Our comparison study indicates that the new model can capture the handover behavior much better than the old one for small-scale cellular networks.

Short Message Service (SMS) is one of the most popular mobile data services. This paper develops a Web-based short message system and analyzes the short message transmission delays based on 40 000 measured data collected from commercial operation. We investigate the distributions of SMS round-trip transmission delays. These distributions are roughly approximated by truncated normal, gamma, log-normal, and Weibull distributions with various parameters, which can be utilized in theoretical analysis. Furthermore, with the nonparametric fitting approach, we use the rejection method to develop random number generators for these delays, which can be used in any simulation models that employ SMS transmission delays.

This paper proposes a radio channel assignment scheme called repacking on demand (RoD) for two-tier wireless local loop (WLL) networks. A two-tier WLL overlays a macrocell with several microcells. When a new call arrives at a two-tier WLL with RoD, if no idle channel is available in both the microcell and the macrocell, repacking is performed (i.e., a call in the macrocell is moved to its corresponding microcell), and then the reclaimed macrocell channel is used to serve the new call. An analytic model is proposed to compute the call blocking probability of the two-tier WLL with repacking. This analytic model is validated against simulation experiments. We prove that the blocking probability is not affected by the call holding time distributions, but is only dependent on the mean of the call holding times. Compared with some previous proposed schemes, RoD has low blocking probability and significantly reduces repacking rate.

In a personal communications service (PCS) network, mobility databases called visitor location registers (VLRs) are utilized to temporarily store the subscription data and the location information for the roaming users. Because of user mobility, it is possible that the VLR is full when a mobile user arrives. Under such a circumstance, the incoming user has no VLR record and thus cannot receive PCS services. This issue is called VLR overflow. To resolve the VLR overflow problem, a VLR record can be selected for replacement when the VLR is full and then the reclaimed storage is used to hold the record of the requesting user. This paper considers the most idle replacement policy to provide services to mobile users without VLR records. In this policy, the record with the longest idle time is selected for replacement. We propose an analytic model to investigate the performance of this replacement policy. The analytic results are validated against simulation experiments. The results indicate that our approach effectively resolves the VLR overflow problem.