Abstract
Information freshness, characterized by age of information (AoI), is important for sensor applications involving timely status updates. In many cases, the wireless signals from one sensor can be received by multiple access points (APs). This paper investigates the average AoI for cooperative APs, in which they can share information through a wired backbone network. We first study a basic backbone-assisted COoperative AP (Co-AP) system where APs share only decoded packets. Experimental results on software-defined radios (SDR) indicate that Co-AP significantly improves the average AoI performance over a single-AP system. Next, we investigate an improved Co-AP system, called Soft-Co-AP. In addition to sharing decoded packets, Soft-Co-AP shares and collects soft information of packets that the APs fail to decode for further joint decoding. A critical issue in Soft-Co-AP is determining the number of quantization bits that represent the soft information (each soft bit) shared over the backbone. While more quantization bits per soft bit improves the joint decoding performance, it leads to higher backbone delay. We experimentally study the average AoI of Soft-Co-AP by evaluating the tradeoff between the backbone delay and the number of quantization bits. SDR experiments show that when the number of sensors is large, Soft-Co-AP further reduces the average AoI by 12% compared with Co-AP. Interestingly, good average AoI performance is usually achieved when the number of quantization bits per soft bit is neither too large nor too small.
Introduction
In modern smart city environments, thousands of sensors—measuring variables like traffic flow, temperature, or air quality—rely on timely status updates. For these applications, the value of data is tied directly to its "freshness," a metric formally defined as the Age of Information (AoI). AoI measures the time elapsed since the last successfully received update was generated at the source.
Wireless channels in dense urban areas are often plagued by shadowing and fading, which can cause a primary access point (AP) to fail to decode a packet. In traditional single-AP systems, such a failure means the information is lost, and the AoI continues to climb until the next successful transmission. However, in many deployments, multiple APs are within range of a single sensor. This paper explores how these neighboring APs can cooperate over a wired Ethernet backbone to significantly improve information freshness.
Cooperative Architectures: Co-AP and Soft-Co-AP
The researchers propose and evaluate three distinct system architectures:
- Single-AP (Baseline): Each sensor communicates with only one primary AP. If decoding fails, the update is lost.
- Co-AP (Cooperative AP): If the primary AP fails but a secondary neighboring AP successfully decodes the packet, the neighbor forwards the fully decoded packet over the Ethernet backbone to the primary AP.
- Soft-Co-AP: This more advanced method addresses cases where neither AP can decode the packet individually. The secondary AP forwards "soft bits"—quantized Log-Likelihood Ratios (LLRs)—to the primary AP. The primary AP then uses Maximum Ratio Combining (MRC) to attempt a joint decode of the combined signals.
The Quantization Tradeoff
A central challenge in Soft-Co-AP is determining the optimal number of quantization bits ($m$) for each soft bit. While increasing $m$ improves the probability of a successful joint decode, it also increases the volume of data sent over the backbone, leading to higher network delay. If this delay exceeds the time until the sensor's next scheduled transmission, the cooperation becomes counterproductive, as the recovered packet would arrive after a newer one has already been sent.
Experimental Methodology
The study utilized a real-world experimental testbed to validate these concepts:
- Hardware: Universal Software Radio Peripheral (USRP) devices served as both sensors and APs.
- Software: The physical layer was implemented using GNU Radio, employing OFDM, BPSK modulation, and Viterbi decoding.
- Data Collection: The team collected 50,000 real physical-layer decoding outcomes across various Signal-to-Noise Ratio (SNR) pairs. They also measured real-world Ethernet backbone delays under high-traffic conditions to accurately model the latency of sharing soft bits.
Key Results and Insights
Experimental results demonstrate that backbone-assisted cooperation provides substantial gains in information freshness. As shown in the performance analysis, even the basic Co-AP system can reduce the average AoI by approximately 50% compared to a single-AP setup when 10 sensors are active in the network.
Further findings include:
- Soft-Co-AP Advantage: In larger networks (e.g., 30 sensors), Soft-Co-AP further reduces the average AoI by 12% compared to basic Co-AP.
- The "Goldilocks" Quantization: The researchers identified that 4 quantization bits per soft bit is often the sweet spot. Using only 1 bit loses too much information, while 8 bits doubles backbone delay without a proportional increase in decoding success.
- Reliability: At a low SNR of 5dB, cooperation increases the packet decoding probability by roughly 30%.
- Spatial Diversity: By leveraging the Ethernet backbone, the system creates a "virtual MIMO" array that is more resilient to local interference and physical obstacles.
Conclusion
This work highlights that for IoT applications, maximizing throughput is not always the primary goal; rather, minimizing the Age of Information is critical for maintaining system timeliness. While the current study assumes a TDMA schedule, future work will explore these cooperative strategies in random-access environments (like Wi-Fi) and investigate more complex cell-free massive MIMO architectures.
Summary updated