Deployment

Deploy scenarios using NEBULA

Advanced Deployment

CPU GPU

Aggregation

FedAvg Krum TrimmedMean Median

Topology

Fully Ring Star

Dataset

MNIST FashionMNIST CIFAR10

Robustness

No Attack Label Flipping Sample Poisoning Model Poisoning

GLL Neuron Inversion Swapping Weights Delayer

Deployment

Scenario Information

Scenario title

Scenario description

Deployment

Deployment type

Federation Architecture

Federation approach

Number of rounds

Network Topology

Topology generation

Dataset

Federated dataset

Dataset type

Partition methods

Parameter setting

Training

Model

Aggregation

Aggregation algorithm

Participants

Logging

Reporting

Individual participants

Advanced Topology

Distance between participants

Advanced Training

Number of Epochs

Robustness

Attack Type

% Malicious nodes

% Samples poisoned by each malicious node

% Poisoned noise

Noise type

Targeted

Target label

Target changed label

Layer

Delay

Flooding Factor

Target Percentage

Target Selection Interval

Starting round

Stopping round

Attack interval

Defense

Enable/Disable Reputation

Reputation Initialization

Weighting Factor

Reputation Metrics

Mobility

Select the default location for the participants

Latitude and longitude

Mobility configuration

Type of mobility

Radius of federation (m)

Scheme of mobility

Each participant moves every (rounds)

% mobile participants

Number of additional participants (deployed during federation)

Schema of deployment