Prospect Pricing in Cognitive Radio Networks

Yingxiang Yang, Leonard T. Park, Narayan B. Mandayam, Ivan Seskar, Arnold Glass, Neha Sinha

TL;DR

This paper introduces "prospect pricing" for cognitive radio networks, a novel approach that accounts for users' psychological biases (specifically, how they perceive risk and uncertainty) when making decisions about service quality. By incorporating Prospect Theory, the authors propose strategies to help Service Providers (SPs) maintain revenue and optimize resource management, even when users tend to undervalue service guarantees.

What Problem Does This Paper Address?

Cognitive Radio Networks (CRNs) aim to efficiently share radio spectrum. Traditional approaches often rely on Expected Utility Theory (EUT) to model user and Service Provider (SP) behavior. However, real-world users don't always act rationally according to EUT, especially when faced with uncertainty about service quality. In CRNs, guaranteeing Quality of Service (QoS) is inherently difficult due to dynamic spectrum availability. This uncertainty can lead users to make decisions that result in underutilized resources and lost revenue for the SP. This paper addresses this gap by modeling user behavior using Prospect Theory, which better captures how people deviate from EUT when making decisions under risk, and developing pricing and resource management strategies that are robust to these psychological biases.

Methods

The core of the paper's methodology involves:

  • Game Theory: A Stackelberg game model is used to represent the interaction between the SP (who moves first by setting prices and allocating bandwidth) and N end-users (who then decide whether to accept the offer).
  • Prospect Theory (PT): Instead of EUT, the authors employ Prospect Theory to model end-user decision-making. This specifically incorporates:
  • Probability Weighting Effect (PWE): Users tend to overweight small probabilities and underweight large ones.
  • Framing Effect (FE): Users evaluate outcomes as gains or losses relative to a reference point (though FE is noted as a topic for future work).
  • Prospect Pricing: A new pricing mechanism is proposed that aims to make the SP's revenue robust to users' PWE.
  • Resource Management Strategies: The paper analyzes several strategies to mitigate revenue loss when users under-weight service guarantees:
  • Bandwidth Reallocation: Dynamically adjusting bandwidth among users.
  • Admission Control: Selecting a subset of users to serve.
  • Bandwidth Expansion/Reduction: Adjusting the total available bandwidth.
  • Rate Control: Offering a different service rate.
  • Analytical Modeling: The paper derives necessary and sufficient conditions for revenue recovery under different scenarios and strategies.
  • Psychophysics Experiments: Human subject studies were conducted to empirically estimate the parameters of the probability weighting function (specifically, Prelec's PWF) based on perceived video QoS.

Datasets

The paper does not explicitly mention using pre-existing datasets for training or evaluation. Instead, it relies on:

  • Simulated Data: Numerical results are generated based on a simulated scenario with 10 users in a cell, using parameters for path loss, shadowing, and Rayleigh fading.
  • Human Subject Data: Preliminary psychophysics experiments were conducted with 23 psychology college students to assess perceived video QoS and estimate the probability weighting function parameters. This data was used to map objective QoS metrics to subjective ratings and derive an empirical probability weighting function.

Key Insights and Ideas

  • Users Don't Always Act Rationally (According to EUT): The paper strongly emphasizes that real-world users deviate from EUT, particularly in how they perceive uncertainty. This deviation can significantly impact SP revenue.
  • Under-weighting Guarantees is Costly: When users under-weight the probability of receiving a guaranteed service quality, they are more likely to reject offers, leading to under-utilization of network resources and revenue loss for the SP.
  • Prospect Pricing as a Solution: The proposed "prospect pricing" mechanism is a key contribution, aiming to make SPs resilient to these user biases by adjusting pricing and resource management strategies.
  • Trade-offs in Resource Management: Different strategies (bandwidth reallocation, admission control, etc.) have varying effectiveness and requirements, often depending on the degree of user bias and available bandwidth.
  • Empirical Validation is Crucial: The inclusion of psychophysics experiments highlights the importance of understanding actual human perception of QoS to accurately model user behavior and parameterize these models.

Key Results

  • Revenue Loss is Significant: When users under-weight service guarantees (i.e., have a skewed probability weighting function), SPs can experience substantial revenue loss if they stick to EUT-based strategies.
  • Bandwidth Reallocation Helps, But Doesn't Fully Recover: Allowing bandwidth reallocation can reduce revenue loss but doesn't fully recover it, especially when strict RRM constraints are maintained.
  • Admission Control's Effectiveness Varies: Admission control can help recover revenue, particularly when the degree of user bias is not too extreme. However, it's less effective for very low values of the probability weighting parameter ($\alpha$).
  • Rate Control is Efficient: Rate control is shown to be an efficient method for revenue recovery, partly because the cost of data rate is higher than bandwidth cost, offering more flexibility.
  • Empirical PWF Parameter: The psychophysics experiments yielded an estimated probability weighting parameter $\alpha \approx 0.585$ for Prelec's PWF, indicating a significant deviation from rational weighting.

Limitations and Future Work

  • Focus on PWE: The paper primarily focuses on the Probability Weighting Effect (PWE) of Prospect Theory, acknowledging that the Framing Effect (FE) is not considered and is a topic for future study.
  • Simplified User Model: While PT is more realistic than EUT, the user benefit functions and cost functions are simplified (e.g., affine cost function).
  • Experimental Scope: The psychophysics experiments were preliminary, involving a limited number of participants and focusing on video QoS. The authors suggest expanding these studies with larger datasets and more detailed mapping techniques.
  • Generalizability of PWF: While Prelec's PWF is used, the authors note that their conclusions can be generalized to other inverse-S-shaped PWFs.
  • Mixed Strategy NE: The analysis of the MUSP game does not consider mixed strategy Nash Equilibria, focusing on pure strategy NE.
  • Future Work: The authors explicitly mention exploring the Framing Effect, conducting more extensive psychophysics studies, and investigating more complex user benefit and cost functions.

Summary updated

Abstract

Advances in cognitive radio networks have primarily focused on the design of spectrally agile radios and novel spectrum sharing techniques that are founded on Expected Utility Theory (EUT). In this paper, we consider the development of novel spectrum sharing algorithms in such networks taking into account human psychological behavior of the end-users, which often deviates from EUT. Specifically, we consider the impact of end-user decision making on pricing and management of radio resources in a cognitive radio enabled network when there is uncertainty in the Quality of Service (QoS) guarantees offered by the Service Provider (SP). Using Prospect Theory (a Nobel-Prize-winning behavioral economic theory that captures human decision making and its deviation from EUT), we design data pricing and channel allocation algorithms for use in cognitive radio networks by formulating a game theoretic analysis of the interplay between the price offerings, bandwidth allocation by the SP and the service choices made by end-users. We show that, when the end-users under-weight the service guarantee, they tend to reject the offer which results in under-utilization of radio resources and revenue loss. We propose prospect pricing, a pricing mechanism that can make the system robust to decision making and improve radio resource management. We present analytical results as well as preliminary human subject studies with video QoS.
BibTeX
@article{yang2015prospect,
  author = {Yingxiang Yang and Leonard T. Park and Narayan B. Mandayam and Ivan Seskar and Arnold Glass and Neha Sinha},
  title = {{Prospect Pricing in Cognitive Radio Networks}},
  journal = {IEEE Trans. Cogn. Commun. Netw.},
  year = {2015},
  url = {https://openreview.net/forum?id=bZAFRI-s7o1},
}
Figure