Thesis submitted in partial fulfilment of the requirements for the degree of Master of Management in Finance and Investment in the faculty of Commerce, Law and Management, Wits Businesss School, University of the Witwatersrand, 2012. / Idiosyncratic considerations relevant to South Africa’s economic and legal framework, as well as determining factors taken from the financial markets and the asset being auctioned are used to critically review the current auction mechanism used by the South African government to borrow funds publicly. A logistic regression with panel data is used in the empirical analysis. The dependent variable has a dichotomous outcome of uniform-price and discriminatory auction mechanisms. Data from 43 different countries over the period 2005 to 2011 are used for the analysis. It was hypothesized that countries with higher uncertainty about the price of their public debt, should use the auction mechanism that reduces under-pricing. Results from the logit regression supported this view. Upon comparing South Africa’s profile with the logit regression results, alongside a review of the literature, it becomes apparent that the proposed model does not provide a definitive answer. However, the model does aid policymakers’ decision on which auction mechanism should be preferred over the other for South Africa.

The emergence of blockchain technology and increasing penetration of distributed energy resources (DERs) have created a new opportunity for peer-to-peer (P2P) energy trading. However, challenges arise in such transactive markets to ensure individual rationality, incentive compatibility, budget balance, and economic efficiency during the trading process. This thesis creates an hour-ahead P2P energy trading network based on the Hyperledger Fabric blockchain and explores a comparative analysis of different auction mechanisms that form the basis of smart contracts. Considered auction mechanisms are discriminatory and uniform k-Double Auction with different k values. This thesis also investigates effects of four consumer and prosumer bidding strategies: random, preference factor, price-only game-theoretic approach, and supply-demand game-theoretic approach. A custom simulation framework that models the behavior of the transactive market is developed. Case studies of a 100-home microgrid at various photovoltaic (PV) penetration levels are presented using typical residential load and PV generation profiles in the metropolitan Washington, D.C. area. Results indicate that regardless of PV penetration levels and employed bidding strategies, discriminatory k-DA can outperform uniform k-DA. Despite so, discriminatory k-DA is more sensitive to market conditions than uniform k-DA. Additionally, results show that the price-only game-theoretic bidding strategy leads to near-ideal economic efficiencies regardless of auction mechanisms and PV penetration levels. / MS

Wireless spectrum is a limited resource that must be used efficiently. It is also a broadcast medium, hence, additional procedures are required to maintain communication over the wireless spectrum private. In this thesis, we investigate three key issues related to efficient use and privacy of wireless spectrum use. First, we propose GAVEL, a truthful short-term auction mechanism that enables efficient use of the wireless spectrum through the licensed shared access model. Second, we propose CPRecycle, an improved Orthogonal Frequency Division Multiplexing (OFDM) receiver that retrieves useful information from the cyclic prefix for interference mitigation thus improving spectral efficiency. Third and finally, we propose WiFi Glass, an attack vector on home WiFi networks to infer private information about home occupants. First we consider, spectrum auctions. Existing short-term spectrum auctions do not satisfy all the features required for a heterogeneous spectrum market. We discover that this is due to the underlying auction format, the sealed bid auction. We propose GAVEL, a truthful auction mechanism, that is based on the ascending bid auction format, that avoids the pitfalls of existing auction mechanisms that are based on the sealed bid auction format. Using extensive simulations we observe that GAVEL can achieve better performance than existing mechanisms. Second, we study the use of cyclic prefix in Orthogonal Frequency Division Multiplexing. The cyclic prefix does contain useful information in the presence of interference. We discover that while the signal of interest is redundant in the cyclic prefix, the interference component varies significantly. We use this insight to design CPRecycle, an improved OFDM receiver that is capable of using the information in the cyclic prefix to mitigate various types of interference. It improves spectral efficiency by decoding packets in the presence of interference. CPRecycle require changes to the OFDM receiver and can be deployed in most networks today. Finally, home WiFi networks are considered private when encryption is enabled using WPA2. However, experiments conducted in real homes, show that the wireless activity on the home network can be used to infer occupancy and activity states such as sleeping and watching television. With this insight, we propose WiFi Glass, an attack vector that can be used to infer occupancy and activity states (limited to three activity classes), using only the passively sniffed WiFi signal from the home environment. Evaluation with real data shows that in most of the cases, only about 15 minutes of sniffed WiFi signal is required to infer private information, highlighting the need for countermeasures.

No description available.

<p dir="ltr">Amidst the rising popularity of digital marketplaces, addressing issues such as non-<br>payment/non-delivery crimes, centralization risks, hacking threats, and the complexity of<br>ownership transfers has become imperative. Many existing studies exploring blockchain<br>technology in digital marketplaces and asset management merely touch upon various application scenarios without establishing a unified platform that ensures trustworthiness and<br>efficiency across the product life cycle. In this thesis, we focus on designing a reliable and efficient e-commerce model to trade various assets. To enhance customer engagement through<br>consensus, we utilize the XGBoost algorithm to identify loyal nodes from the platform entities pool. Alongside appointed nodes, these loyal nodes actively participate in the consensus<br>process. The consensus algorithm guarantees that all involved nodes reach an agreement on<br>the blockchain’s current state. We introduce a novel consensus mechanism named Modified-<br>Practical Byzantine Fault Tolerance (M-PBFT), derived from the Practical Byzantine Fault<br>Tolerance (PBFT) protocol to minimize communication overhead and improve overall efficiency. The modifications primarily target the leader election process and the communication<br>protocols between leader and follower nodes within the PBFT consensus framework.</p><p dir="ltr"><br>In the domain of tangible assets, our primary objective is to elevate trust among various<br>stakeholders and bolster the reputation of sellers. As a result, we aim to validate secondhand<br>products and their descriptions provided by the sellers before the secondhand products are<br>exchanged. This validation process also holds various entities accountable for their actions.<br>We employ validators based on their location and qualifications to validate the products’<br>descriptions and generate validation certificates for the products, which are then securely<br>recorded on the blockchain. To incentivize the participation of validator nodes and up-<br>hold honest validation of product quality, we introduce an incentive mechanism leveraging<br>Stackelberg game theory.</p><p dir="ltr"><br>On the other hand, for optimizing intangible assets management, we employ Non-Fungible<br>Tokens (NFT) technology to tokenize these assets. This approach enhances traceability of<br>ownership, transactions, and historical data, while also automating processes like dividend<br>distributions, royalty payments, and ownership transfers through smart contracts. Initially,<br>sellers mint NFTs and utilize the InterPlanetary File System (IPFS) to store the files related<br>to NFTs, NFT metadata, or both since IPFS provides resilience and decentralized storage solutions to our network. The data stored in IPFS is encrypted for security purposes.<br>Further, to aid sellers in pricing their NFTs efficiently, we employ the Stackelberg mechanism. Furthermore, to achieve finer access control in NFTs containing sensitive data and<br>increase sellers’ profits, we propose a Popularity-based Adaptive NFT Management Scheme<br>(PANMS) utilizing Reinforcement Learning (RL). To facilitate prompt and effective asset<br>sales, we design a smart contract-powered auction mechanism.</p><p dir="ltr"><br>Also, to enhance data recording and event response efficiency, we introduce a weighted<br>L-H index algorithm and transaction prioritization features in the network. The weighted<br>L-H index algorithm determines efficient nodes to broadcast transactions. Transaction prior-<br>itization prioritizes certain transactions such as payments, verdicts during conflicts between<br>sellers and validators, and validation reports to improve the efficiency of the platform. Simulation experiments are conducted to demonstrate the accuracy and efficiency of our proposed<br>schemes.<br></p>