Environment-Adaptive Multiple Access for Distributed V2X Network: A Reinforcement Learning Framework

Dedicated short-range communications (DSRC) is one of the key technologies enabling safety-critical applications for intelligent transportation system (ITS). Considering the significance of such safety-of-life applications, it is of utmost importance to guarantee reliable delivery of basic safety messages (BSMs). However, in accordance with a V2X network being inherently dynamic in key aspects such as vehicle density and velocity, the networking behavior of a DSRC system is usually highly complicated to analyze. In addition, the United States Federal Communications Commission (US FCC) recently proposed the so-called “5.9 GHz band innovation,”which includes a plan to reduce bandwidth for DSRC to 10 MHz at best from 75 MHz. Motivated from these challenges, the necessity of “lightening”load of a DSRC network has become essential to keep safety-related operations from performance deterioration. To this end, this paper proposes a protocol that prioritizes transmission of a BSM from a vehicle according to the level of accident risk of the vehicle. The proposed protocol uses the distance of a vehicle from a danger source as the metric to determine the priority for transmission. Our results show that this protocol effectively prioritizes the transmission opportunity for dangerous vehicles, and hence results in higher performance in terms of key metrics-i.e., average latency, packet delivery rate (PDR), and inter-reception time (IRT).

A Contextual Multi-Armed Bandit approach for NDN forwarding

Sophisticated Spectrum Sharing Analysis: The Case of the 5.9 GHz Band

Vehicle-to-Everything (V2X) communication is an emerging technology, a vital enabler in realizing the concepts of smart cities, which helps in autonomous driving and intelligent transportation systems. This paper focuses on the performance comparison of Wi-Fi, Dedicated Short Range Communications (DSRC), and modified DSRC when used in a V2X communication network. The communication scenario considered includes message passing for V2X applications like emergency brake assist, intersection collision warning, lane departure warning, signal status display in-vehicle, school zone warning, and, parking slot status display as per the SAE J2735 standard. The communication network under various operational scenarios is implemented in NetSim, a licensed network simulator. The performance evaluation is done based on metrics like packet delivery ratio (PDR), MAC layer, and physical layer processing time, link throughput, and delay. Performance analysis based on unicast messages of DSRC and Wi-Fi shows that Wi-Fi gives better performance. A modified version of DSRC using variants of Wi-Fi can improve the performance of DSRC as evident from the performance analysis of modified DSRC.

The 5.9 GHz spectrum band is proposed for vehicular communications using Dedicated Short Range Communications (DSRC), but this band may need to be shared with unlicensed Wi-Fi devices. A recent Federal Communications Commission (FCC)’s Notice of Proposed Rulemaking (NPRM) outlines two interference mitigation techniques for spectrum coexistence: Detect and Vacate and Re-channelization. A major technical challenge of Re-channelization is that DSRC may experience harmful or a reduced functionality because of adjacent channel interference from Wi-Fi transmitters. We therefore conducted Wi-Fi/DSRC adjacent channel interference experiments to evaluate the significance of Wi-Fi signals adjacent to DSRC transmissions. Our measurements show a significant degradation of DSRC performance from 802.11ac adjacent channel interference only for certain scenarios where the distances between the Wi-Fi transmitter to DSRC receiver is 15 m or below and the distances between the DSRC transmitter to receiver is 300m or higher. However, there is no severe effect when the DSRC transmit power is at the standard 33 dBm level, even considering the recommended fading margin of 5-10 dB.

Flooding mitigation through safe & trustworthy reinforcement learning

The dedicated short range communication (DSRC) is a spring-up technique that pushes intelligent transportation systems (ITS) into our habitual life. Intelligent transportation services involves safer and more coordinated way of using transport networks. Routine broadcast of vehicular safety messages is one of the perspectives of intelligent transportation systems. Developed countries such as Japan, USA and Europe customarily adopts FM0 and Manchester codes for DSRC in intelligent transportation systems to reach DC balance. Nonetheless, the coding diversity between the two codes restraints a fully reused VLSI architecture for the two coding methods. To unbound the limitation, the paper proposes similarity oriented logic simplification (SOLS) technique. The pressing feature of this technique is that it augments the hardware utilization rate (HUR) from 57.14% to 100% for the two encoding methods. The proposed architecture can fully support the DSRC standards of the developed countries. To enhance the efficiency of DSRC applications, the paper also proposes error detection and error correction through hamming codes. Hamming codes are used for error control. This paper ensures an error-free DSRC communication system along with area and power efficient encoder design. The paper is simulated in ModelSim 6.3f using VHDL. The design is implemented in Spartan 6 XC6SLX16.

The exploration is built upon Delphi’s, Nissan’s, Cohda Wireless’ and Savari’s experiences in Asia, Europe and U.S.A. It describes and derives lessons from all four companies’ contributions in projects such as SMARTWAY in Japan, Drive C2X and in Europe, as well as the Connected Vehicle Safety Pilot in the U.S.A. All the above programs were implemented by means of the Dedicated Short Range Communication (DSRC) technology in the SHF spectrum based on the IEEE 802.11p/Wireless Access in Vehicular Environments (WAVE) standard. The study is supplemented with insights regarding complementary technologies such DSRC in the lower UHF frequency band (i.e. 700 MHz) as well as a V2X implementation through the 4G LTE (Long Term Evolution) cellular telecommunication technology. This paper addresses issues regarding the physical layer (PHY) of the DSRC system. The combination of the delay profile caused by multipath propagation along with the motion-based Doppler spread leads to time and frequency dispersion. This limits the number of bytes acceptable for reliable communication or requires a solution at the receiver end. The analysis of the Doppler spread shows that DSRC implemented at 700 MHz is more immune from data packet length issues as opposed to 5 GHz DSRC. On the other hand, 700 MHz DSRC exhibits a much longer delay spread. Thus, guard time interval specified in ASTM E2213-03 cannot be applied as is to 700 MHz DSRC. This paper refers to the German project CoCarX and the Japanese SKY for pedestrian for studying the feasibility a V2X system built on the 4G/LTE technology and its infrastructure. It provides on a vision for an accelerated V2X deployment based on a heterogeneous system. Last, we recommend the ITS stakeholders to carry out extensive research and validation works on DSRC capacity for ensuring a large scale deployment.

Most works on Multi-Armed Bandits focus the evaluations of their methods on a global accuracy performance metric. In the case of Contextual Multi-Armed Bandit (CMAB), the existing algorithms claim to eventually provide full personalization. This suggests that their global accuracy metric should reflect each user's individual accuracy. In order to verify this, we consider a novel approach of CMAB assessment focused on the evaluation of individual accuracy and compare it to global accuracy. Based on the results of this comparison highlighting some users far from the average global accuracy we propose a heuristic, Sliding Window LinUCB (SW-LinUCB), aiming at decreasing the amount of very unsatisfied users. SW-LinUCB is an adaptation of the original LinUCB CMAB algorithm combined with a diversification mechanism. It acts as the original LinUCB but penalizes arms which are pulled too frequently in order to play them evenly or fairly among optimum. Our approach is motivated by the requirements of recommender systems, which must reach a good global accuracy and should equally distribute it among individuals. We experiment and discuss the benefits and losses of the proposed method on several real-world datasets.

This paper presents the design and implementation of a learning controller for the Automatic Generation Control (AGC) in power systems based on a reinforcement learning (RL) framework. In contrast to the recent RL scheme for AGC proposed by us, the present method permits handling of power system variables such as Area Control Error (ACE) and deviations from scheduled frequency and tie-line flows as continuous variables. (In the earlier scheme, these variables have to be quantized into finitely many levels). The optimal control law is arrived at in the RL framework by making use of Q-learning strategy. Since the state variables are continuous, we propose the use of Radial Basis Function (RBF) neural networks to compute the Q-values for a given input state. Since, in this application we cannot provide training data appropriate for the standard supervised learning framework, a reinforcement learning algorithm is employed to train the RBF network. We also employ a novel exploration strategy, based on a Learning Automata algorithm, for generating training samples during Q-learning. The proposed scheme, in addition to being simple to implement, inherits all the attractive features of an RL scheme such as model independent design, flexibility in control objective specification, robustness etc. Two implementations of the proposed approach are presented. Through simulation studies the attractiveness of this approach is demonstrated.

Nitrous oxide (N2O) emissions from agriculture are rising due to increased fertilizer use and intensive farming, posing a major challenge for climate mitigation. This study introduces a novel reinforcement learning (RL) framework to optimize farm management strategies that balance crop productivity with environmental impact, particularly N2O emissions. By modeling agricultural decision-making as a partially observable Markov decision process (POMDP), the framework accounts for uncertainties in environmental conditions and observational data. The approach integrates deep Q-learning with recurrent neural networks (RNNs) to train adaptive agents within a simulated farming environment. A Probabilistic Deep Learning (PDL) model was developed to estimate N2O emissions, achieving a high Prediction Interval Coverage Probability (PICP) of 0.937 within a 95% confidence interval on the available dataset. While the PDL model’s generalizability is currently constrained by the limited observational data, the RL framework itself is designed for broad applicability, capable of extending to diverse agricultural practices and environmental conditions. Results demonstrate that RL agents reduce N2O emissions without compromising yields, even under climatic variability. The framework’s flexibility allows for future integration of expanded datasets or alternative emission models, ensuring scalability as more field data becomes available. This work highlights the potential of artificial intelligence to advance climate-smart agriculture by simultaneously addressing productivity and sustainability goals in dynamic real-world settings.

Vehicular mesh networks could be an important new way to provide Internet access in urban areas using dedicated short range communications (DSRC). In some circumstances, DSRC technology is more cost-effective than expanding the capacity of cellular networks. We determine what those circumstances are by combining our simulation model with data collected from an actual vehicular network that is operating in Portugal. We use the model to estimate how much Internet traffic can be offloaded to vehicular networks that would otherwise be carried by cellular networks, under a variety of conditions. We use offloaded traffic to estimate the benefits of cost savings of reduced cellular infrastructure due to offload, and the cost of the DSRC vehicular network to carry that traffic. Then, we determine when benefit exceeds cost. We find that the benefits from the Internet traffic alone are not enough to justify a universal mandate to deploy DSRC in all vehicles, i.e., the benefits of Internet access alone are less than total costs. However, the majority of DSRC-related costs must be incurred anyway if safety is to be enhanced. Thus, soon after a mandate to put DSRC in new vehicles becomes effective, the benefits of Internet access through vehicular networks in densely populated areas would be significantly greater than the remaining costs, which are the costs of roadside infrastructure that can serve as a gateway to the Internet. Moreover, the benefit of Internet access would exceed DSRC infrastructure cost in regions with lower and lower population densities over time.

Dedicated Short Range Communication (DSRC) is a key enabling technology for the next generation of communication-based safety applications. Distance measurement based on RSSI, featuring low communication overhead and low complexity, is widely applied in the range-based localization of DSRC communication. However, the uncertainty factor and range error of shadowing model vary under different circumstances. Furthermore, the drawback of positioning error will affect its priority in geocasting or multi-hop of DSRC signals. This paper presents an operational scheme of parameters contribution design by utilizing Taguchi method, and the experiment inference is redesigned to compensate circumstance factors for shadowing model using genetic algorithm in the proving ground. The genetic algorithm provides an advanced solution to the circumstance factor and measured uncertainty using minimal absolute errors. These results may serve as benchmarks of parameters design for future DSRC channel communication.

In this paper, the theoretical as well as experimental results of BER characteristics of three different modulation schemes, ASK, FSK and BPSK, in a multi-path Rician channel are addressed. These BER characteristics are analyzed as a function of E b / N o and the power ratio of the line of sight (LOS) component to the Rayleigh scattered component The theoretical as well as computer simulation results shows the ASK is the most suitable modulation scheme for the dedicated short range communication (DSRC) in terms of implemental cost and system complexity. The decision feedback equalizer is proved to be very effective in canceling the multi-path interference in the DSRC channel environment. The simulation result of the equalized ASK, reveals the performance enhancement achievable with decision feedback (DFE) equalizer for the first generation DSRC system. The multi-ray DSRC channel model is also provided to predict the received carrier power and fluctuation, which are quite dependent on the surroundings of a cell.

실시간 자료를 반영한 통행시간 예측 기법은 다양하지만 관련 연구 검토 결과 과거이력데이터가 충분하다면 타 모형에 비해 K 최대근접이웃(K-Nearest Neighbors)의 정확도가 우수하므로 본 연구에서는 이에 대한 적용 방법 도출 및 가능성 평가를 목적으로 한다. 본 연구에서는 KNN의 입력 자료로 TCS 교통량 및 DSRC 구간통행시간의 실시간 및 과거 이력자료, 경로통행시간 이력자료를 활용하였다. 통행시간 예측치는 TCS 교통량 및 DSRC 구간통행시간의 실시간 자료와 유사한 경로통행시간을 탐색한 후 이를 가중평균하여 산출하였다. 예측 기법을 적용한 결과 DSRC 구간통행시간의 가중치가 증가할수록 정확도는 증가하였으며, 이는 실시간 교통상황 변화를 DSRC 구간통행시간이 잘 반영하기 때문이다. 그러나 TCS 교통량을 기반으로 한 경우 역시 정확도의 차이가 크지 않으며, 변화 추이도 유사하게 나타났다. 이러한 결과를 볼 때 향후 대용량의 과거이력자료가 축적될 경우 예측오차는 더욱 감소될 것으로 기대된다. There are various methodologies to forecast the travel time using real-time data but the K-nearest neighborhood (KNN) method in general is regarded as the most one in forecasting when there are enough historical data. The objective of this study is to evaluate applicability of KNN method. In this study, real-time and historical data of toll collection system (TCS) traffic flow and the dedicated short range communication (DSRC) link travel time, and the historical path travel time data are used as input data for KNN approach. The proposed method investigates the path travel time which is the nearest to TCS traffic flow and DSRC link travel time from real-time and historical data, then it calculates the predicted path travel time using weight average method. The results show that accuracy increased when weighted value of DSRC link travel time increases. Moreover the trend of forecasted and real travel times are similar. In addition, the error in forecasted travel time could be further reduced when more historical data could be available in the future database.