-
From Unstructured Communication to Intelligent RAG: Multi-Agent Automation for Supply Chain Knowledge Bases
Authors:
Yao Zhang,
Zaixi Shang,
Silpan Patel,
Mikel Zuniga
Abstract:
Supply chain operations generate vast amounts of operational data; however, critical knowledge such as system usage practices, troubleshooting workflows, and resolution techniques often remains buried within unstructured communications like support tickets, emails, and chat logs. While RAG systems aim to leverage such communications as a knowledge base, their effectiveness is limited by raw data c…
▽ More
Supply chain operations generate vast amounts of operational data; however, critical knowledge such as system usage practices, troubleshooting workflows, and resolution techniques often remains buried within unstructured communications like support tickets, emails, and chat logs. While RAG systems aim to leverage such communications as a knowledge base, their effectiveness is limited by raw data challenges: support tickets are typically noisy, inconsistent, and incomplete, making direct retrieval suboptimal. Unlike existing RAG approaches that focus on runtime optimization, we introduce a novel offline-first methodology that transforms these communications into a structured knowledge base. Our key innovation is a LLMs-based multi-agent system orchestrating three specialized agents: Category Discovery for taxonomy creation, Categorization for ticket grouping, and Knowledge Synthesis for article generation. Applying our methodology to real-world support tickets with resolution notes and comments, our system creates a compact knowledge base - reducing total volume to just 3.4% of original ticket data while improving quality. Experiments demonstrate that our prebuilt knowledge base in RAG systems significantly outperforms traditional RAG implementations (48.74% vs. 38.60% helpful answers) and achieves a 77.4% reduction in unhelpful responses. By automating institutional knowledge capture that typically remains siloed in experts' heads, our solution translates to substantial operational efficiency: reducing support workload, accelerating resolution times, and creating self-improving systems that automatically resolve approximately 50% of future supply chain tickets. Our approach addresses a key gap in knowledge management by transforming transient communications into structured, reusable knowledge through intelligent offline processing rather than latency-inducing runtime architectures.
△ Less
Submitted 20 June, 2025;
originally announced June 2025.
-
Testing an entropy estimator related to the dynamical state of galaxy clusters
Authors:
J. M. Zúniga,
C. A. Caretta,
A. P. González,
E. García-Manzanárez
Abstract:
We propose the entropy estimator $H_Z$, calculated from global dynamical parameters, in an attempt to capture the degree of evolution of galaxy systems. We assume that the observed (spatial and velocity) distributions of member galaxies in these systems evolve over time towards states of higher dynamical relaxation (higher entropy), becoming more random and homogeneous in virial equilibrium. Thus,…
▽ More
We propose the entropy estimator $H_Z$, calculated from global dynamical parameters, in an attempt to capture the degree of evolution of galaxy systems. We assume that the observed (spatial and velocity) distributions of member galaxies in these systems evolve over time towards states of higher dynamical relaxation (higher entropy), becoming more random and homogeneous in virial equilibrium. Thus, the $H_Z$-entropy should correspond to the gravitacional assembly state of the systems. This was tested in a sample of 70 well sampled clusters in the Local Universe whose gravitational assembly state, classified from optical and X-ray analysis of substructures, shows clear statistical correlation with $H_Z$. This estimator was also tested on a sample of clusters (halos) from the IllustrisTNG simulations, obtaining results in agreement with the observational ones.
△ Less
Submitted 7 March, 2024;
originally announced March 2024.
-
CardioID: Mitigating the Effects of Irregular Cardiac Signals for Biometric Identification
Authors:
Weizheng Wang,
Marco Zuniga,
Qing Wang
Abstract:
Cardiac patterns are being used to obtain hard-to-forge biometric signatures and have led to high accuracy in state-of-the-art (SoA) identification applications. However, this performance is obtained under controlled scenarios where cardiac signals maintain a relatively uniform pattern, facilitating the identification process. In this work, we analyze cardiac signals collected in more realistic (u…
▽ More
Cardiac patterns are being used to obtain hard-to-forge biometric signatures and have led to high accuracy in state-of-the-art (SoA) identification applications. However, this performance is obtained under controlled scenarios where cardiac signals maintain a relatively uniform pattern, facilitating the identification process. In this work, we analyze cardiac signals collected in more realistic (uncontrolled) scenarios and show that their high signal variability (i.e., irregularity) makes it harder to obtain stable and distinct user features. Furthermore, SoA usually fails to identify specific groups of users, rendering existing identification methods futile in uncontrolled scenarios. To solve these problems, we propose a framework with three novel properties. First, we design an adaptive method that achieves stable and distinct features by tailoring the filtering spectrum to each user. Second, we show that users can have multiple cardiac morphologies, offering us a much bigger pool of cardiac signals and users compared to SoA. Third, we overcome other distortion effects present in authentication applications with a multi-cluster approach and the Mahalanobis distance. Our evaluation shows that the average balanced accuracy (BAC) of SoA drops from above 90% in controlled scenarios to 75% in uncontrolled ones, while our method maintains an average BAC above 90% in uncontrolled scenarios.
△ Less
Submitted 30 March, 2022;
originally announced March 2022.
-
The Cascading Metric Tree
Authors:
Jeffrey Uhlmann,
Miguel R. Zuniga
Abstract:
This paper presents the Cascaded Metric Tree (CMT) for efficient satisfaction of metric search queries over a dataset of N objects. It provides extra information that permits query algorithms to exploit all distance calculations performed along each path in the tree for pruning purposes. In addition to improving standard metric range (ball) query algorithms, we present a new algorithm for exploiti…
▽ More
This paper presents the Cascaded Metric Tree (CMT) for efficient satisfaction of metric search queries over a dataset of N objects. It provides extra information that permits query algorithms to exploit all distance calculations performed along each path in the tree for pruning purposes. In addition to improving standard metric range (ball) query algorithms, we present a new algorithm for exploiting the CMT cascaded information to achieve near-optimal performance for k-nearest neighbor (kNN) queries. We demonstrate the performance advantage of CMT over classical metric search structures on synthetic datasets of up to 10 million objects and on the 564K Swiss-Prot protein sequence dataset containing over $200$ million amino acids. As a supplement to the paper, we provide reference implementations of the empirically-examined algorithms to encourage improvements and further applications of CMT to practical scientific and engineering problems
△ Less
Submitted 20 December, 2021;
originally announced December 2021.
-
BEH: Indoor Batteryless BLE Beacons using RF Energy Harvesting for Internet of Things
Authors:
Qingzhi Liu,
Wieger IJntema,
Anass Drif,
Przemysław Pawełczak,
Marco Zuniga
Abstract:
Indoor beacon system is a cornerstone of internet of things. Until now, most of the research effort has focused on achieving various applications based on beacon communication. However, the power consumption on beacon devices becomes the bottleneck for the large-scale deployments of indoor networks. On one hand, the size of beacon devices is required to be small enough for easy deployment, which f…
▽ More
Indoor beacon system is a cornerstone of internet of things. Until now, most of the research effort has focused on achieving various applications based on beacon communication. However, the power consumption on beacon devices becomes the bottleneck for the large-scale deployments of indoor networks. On one hand, the size of beacon devices is required to be small enough for easy deployment, which further limits the size of battery. On the other hand, replacing the batteries of the beacon devices could lead to high maintenance costs. It is important to provide them with full energy autonomy. To tackle this problem we propose BEH: an indoor beacon system aimed at operating perpetually without batteries. Our contributions are twofold. Firstly, we propose four methods to increase the utilization efficiency of harvested RF energy in the beacon system, by which the energy consumption level becomes low enough to fit within the energy harvesting budget. Secondly, we implement BEH using Bluetooth Low Energy (BLE) and Radio Frequency (RF) energy harvesting devices, and test BEH extensively in a laboratory environment. Our test results show that BEH can enable perpetual lifetime operation of mobile and static BLE beacon devices in a (16 m$^{\text{2}}$) room. The average value of packet reception rate achieves more than 99% in the best case.
△ Less
Submitted 8 November, 2019;
originally announced November 2019.
-
WiPLoc: Perpetual Indoor Localization with RF Wireless Power Transfer
Authors:
Qingzhi Liu,
Wieger IJntema,
Anass Drif,
Przemysław Pawełczak,
Marco Zuniga
Abstract:
Indoor localization is a cornerstone of mobile services. Until now most of the research effort has focused on achieving sub-meter localization accuracy because many mobile applications depend on precise localization measurements. In some scenarios, however, it is required to trade location accuracy for system maintainability. For example, in large-scale deployments of indoor networks, such as item…
▽ More
Indoor localization is a cornerstone of mobile services. Until now most of the research effort has focused on achieving sub-meter localization accuracy because many mobile applications depend on precise localization measurements. In some scenarios, however, it is required to trade location accuracy for system maintainability. For example, in large-scale deployments of indoor networks, such as item-monitoring in smart buildings, attaining room-level localization accuracy may be sufficient, but replacing the batteries of the devices used for localization could lead to high operational costs. As indoor localization systems grow in popularity it will be important to provide them with full energy autonomy. To tackle this problem we propose WiPLoc: an indoor localization system aimed at operating perpetually without batteries. Our contributions are twofold. First, we propose a novel localization method that exploits capture effect and orthogonal codes to operate at energy levels that are low enough to operate within the energy budget provided by long-range wireless power transmission. Second, we implement WiPLoc using off-the-shelf components and test it extensively in a laboratory environment. Our test results show that with WiPLoc one wireless charger per (16 m$^{\text{2}}$) room can enable perpetual lifetime operation of mobile objects requiring localization with an average accuracy of almost 90%.
△ Less
Submitted 5 December, 2017;
originally announced December 2017.
-
Passive Sensing and Communication Using Visible Light: Taxonomy, Challenges and Opportunities
Authors:
Qing Wang,
Marco Zuniga
Abstract:
For more than a century, artificial lighting has served mainly for illumination. Only recently, we start to transform our lighting infrastructure to provide new services such as indoor localization and network connectivity. These innovative advancements rely on two key requirements: the ability to modulate light sources (for data transmission) and the presence of photodetectors on objects (for dat…
▽ More
For more than a century, artificial lighting has served mainly for illumination. Only recently, we start to transform our lighting infrastructure to provide new services such as indoor localization and network connectivity. These innovative advancements rely on two key requirements: the ability to modulate light sources (for data transmission) and the presence of photodetectors on objects (for data reception). But not all lights can be modulated and most objects do not have photodetectors. To overcome these limitations, researchers are developing novel sensing and communication methods that exploit passive light sources, such as the sun, and that leverage the external surfaces of objects, such as fingers and car roofs, to create a new generation of cyber-physical systems based on visible light. In this article we propose a taxonomy to analyze these novel contributions. Our taxonomy allows us to identify the overarching principles, challenges and opportunities of this new rising area.
△ Less
Submitted 5 April, 2017;
originally announced April 2017.
-
Passive Communication with Ambient Light
Authors:
Qing Wang,
Marco Zuniga,
Domenico Giustiniano
Abstract:
In this work, we propose a new communication system for illuminated areas, indoors and outdoors. Light sources in our environments -such as light bulbs or even the sun- are our signal emitters, but we do not modulate data at the light source. We instead propose that the environment itself modulates the ambient light signals: if mobile elements 'wear' patterns consisting of distinctive reflecting s…
▽ More
In this work, we propose a new communication system for illuminated areas, indoors and outdoors. Light sources in our environments -such as light bulbs or even the sun- are our signal emitters, but we do not modulate data at the light source. We instead propose that the environment itself modulates the ambient light signals: if mobile elements 'wear' patterns consisting of distinctive reflecting surfaces, single photodiode could decode the disturbed light signals to read passive information. Achieving this vision requires a deep understanding of a new type of communication channel. Many parameters can affect the performance of passive communication based on visible light: the size of reflective surfaces, the surrounding light intensity, the speed of mobile objects, the field-of-view of the receiver, to name a few. In this paper, we present our vision for a passive communication channel with visible light, the design challenges and the evaluation of an outdoor application where our receiver decodes information from a car moving at 18 km/h.
△ Less
Submitted 6 January, 2017; v1 submitted 14 December, 2015;
originally announced December 2015.
-
Jam-X: Wireless Agreement under Interference
Authors:
Carlo Alberto Boano,
Kay Römer,
Marco Antonio Zúñiga,
Thiemo Voigt
Abstract:
Wireless low-power transceivers used in sensor networks such as IEEE 802.15.4 typically operate in unlicensed frequency bands that are subject to external interference from devices transmitting at much higher power. Communication protocols should therefore be designed to be robust against such interference. A critical building block of many protocols at all layers is agreement on a piece of inform…
▽ More
Wireless low-power transceivers used in sensor networks such as IEEE 802.15.4 typically operate in unlicensed frequency bands that are subject to external interference from devices transmitting at much higher power. Communication protocols should therefore be designed to be robust against such interference. A critical building block of many protocols at all layers is agreement on a piece of information among a set of nodes. At the MAC layer, nodes may need to agree on a new time slot or frequency channel; at the application layer nodes may need to agree on handing over a leader role from one node to another. Message loss caused by interference may break agreement in two different ways: none of the nodes use the new information (time slot, channel, leader) and stick with the previous assignment, or - even worse - some nodes use the new information and some do not. This may lead to reduced performance or failures. In this paper we investigate the problem of agreement under interference and point out the limitations of the traditional message-based n-way handshake approach. We propose novel protocols that use jamming instead of message transmissions and show that they outperform the n-way handshake in terms of agreement probability, energy consumption, and time-to-completion both in the unicast case (two neighboring nodes agree) as well as in the broadcast case (any number of neighboring nodes agree).
△ Less
Submitted 22 February, 2012;
originally announced February 2012.
